(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Research and Technology 1996"



Research and Technology 

T9T6 



Ames Research Center 



National Aeronautics 
and Space Administration 

Ames Research Center 

Moffett Field, California 



NASA TM-l 12195 



Foreword 



The mission of NASA Ames Research Center is to research, develop, verify, and transfer advanced 
aeronautics, space, and related technologies; to advance and communicate scenff.c knowledge 
and understanding of the universe, the solar system, and the Earth; and to enable the development 
"space for human enterprise. Emphasis is placed on information systems ^-'^"f ^-es for -o- 
nautL and space applications; on aviation operations systems; and on the d.scpime of astrob.ol- 
ogy, the study of life in the universe encompassing the Earth, space, and life sciences. 

This report highlights the challenging work accomplished during fiscal year 1996 by Ames research 
scienti^s, engineers, and technologists. It discusses research and technologies that enable he 
information Age, that expand the frontiers of knowledge for aeronautics and space, and that help 
o maintain U S. leadership in aeronautics and space research and technology development. The 
accomplishments span the range of goals of NASA's four Strategic Enterprises Aeronautics and 
Space Transportation Technology, Space Science, Human Exploration and Development of Space, 
and Mission to Planet Earth. 

The primary purpose of this report is to communicate knowledge-to inform our stakeholders, 
customers, and partners, and the people of the United States about the scope and diversity of 
Ames' mi;sion, the nature of Ames' research and technology activities, and the stimulating chal- 
lenges ahead. The accomplishments cited illustrate the contributions ^h^* /^mes is making to 
improve the quality of life for our citizens and the economic position of the United States in the 
world marketplace. 
For further information on Ames research and technology projects, please contact the person 

as^'the" point of contact at the end of each article. An electronic version of this report is available at 

URL http//jit.arc,nasa,gov/atrs/index.html. 



^/W^ ^' ^-^ 



Henry McDonald 
Director 



Cost t: v t s 



Aeronautics Enterprise 

Overview 1 

Global Civil Aviation/Safety 

Aviation Performance Measuring System 3 

Irving C, Statler 

Aviation Safety Reporting System 4 

Linda ), Connell 

Crew Activity Tracking System 5 

Todd ). Callantine 

Study of Line Oriented Flight Training 6 

Key R. Dismukes 

Measuring Air Traffic Complexity 7 

Irene V. Laudeman, Connie Brasil, Robert Branstrom 

Operational Interventions to Human Error in Aircraft Maintenance 8 

Barbara G. Kanki, Vicki Dulchinos 

Aircraft Separation Risk Model 9 

Mary M. Connors 

The Final Approach Spacing Tool 1 1 

Tom Davis 

Air/Ground Integration 11 

R. Slattery 

The Traffic Management Advisor 12 

Harry N. Swenson 

Surface Movement Advisor 13 

Brian J. Glass 

Simplified Vision Models for Display Quality Assessment 14 

Al Ahumada 

Perceptually Tuned Visual Simulation 15 

Mary K. Kaiser 

Entropy Masking in Visual Displays 17 

Andrew B. Watson 

Vertical Motion Simulator Advanced Simulator Network 18 

William B. Cleveland 

intelligent Aircraft Control System 19 

Charles C. jorgensen 



in 



Aeronautics Enterprise (continued) 

Global Civil Aviation/Affordability 

Facilitating User Route Preferences in En Route Airspace 20 

Bob Vivona, Mark Baltin, Steve Green, Ralph Bach, Dave McNally 

Conflict Probability Estimation for Free Flight 21 

Russell A. Paielli, Heinz Erzberger 

Conflict Prediction Algorithms: Initial Field Test 23 

Dave McNally, Bob Vivona, Karl Bilimoria, Gerd Kanning, Steve Green, Ralph Bach, 

Allan McCrary, Ed Levi^is 
Assessing Controller Performance Under Simulated Free-Flight Conditions 24 

Roger Remington, James Johnston, Eric Ruthruff, Maria Romera 

Evaluating Cockpit Display of Traffic Information Displays and Route Assessment Tools 

for a Free Flight Environment 25 

Vernol Battiste, Walter W. Johnson 

Cockpit Displays for Low-Visibility Taxiing 27 

David Foyle, Elizabeth M. Wenzel, Durand R. Begault 

Multi-Sensor Image Registration 29 

Misha Pavel, Al Ahumada, Barbara Sweet 

Flow Visualization of a Full-Scale Rotor in Hover 29 

Benton H. Lau, Alan j. Wadcock, Gloria K. Yamauchi 

Skin-Friction Measurements on a Hovering Rotor 30 

Alan ). Wadcock, Gloria K. Yamauchi 

Rotor Data Correlation 31 

Randall Peterson 

Canard RotorAVing Hover Test 32 

Stephen Swanson, John Madden 

In-flight Dynamic Stall Research 34 

Robert Kufeld, William Bousman 

Stall Control of Helicopter Rotors 34 

Khanh Q, Nguyen 

Apache AH-64D Flight Test Predictions 35 

Earl P. N. Duque 

Rotor- Wake/Fuselage Interaction 36 

Paul M. Stremel 
Navier-Stokes Simulation of High-Lift Aerodynamics 37 

Karlin Roth, Stuart Rogers 

Lift-Jet Effects on Powered-Lift STOVL Model 38 

Karlin Roth 
Vortex Core Detection for Computational Grid Refinement 38 

David Kenwright 



IV 



C ^ T E \ T S 



Aeronautics Enterprise (continued) 

Global Civil Aviation/Affordability (continued) 

Wingtip Vortex Flows :jq 

Jennifer Dacles-Mariani, Dochan Kwak 

Transonic Overset Potential Solver ^q 

Terry Hoist 

Real-Time Particle Tracing in Time-Varying Flows 4^ 

David Kenwrighl David Kao 

NASA Metacenter ^2 

Mary Hultquist 

Portable Batch System .o 

David Tweten 

Developing a Cluster Computer from Workstations 43 

Reese L, Sorenson 

Planar Doppler Velocimetry Using Pulsed Lasers 44 

Robert L. McKenzie 

Pressure-Sensitive Paint and Photogrammetry for Aeroelastic Experiments 45 

Edward T. Schairer, Lawrence A. Hand 

Visualizing Wind-Tunnel Experimental Data 47 

Samuel P. Uselton, Glenn Deardorff, Leslie Keely, Yinsyi Hung, Arsi Vaziri 

Surface Tension Effects on Skin Friction Measurements 48 

G. Zilliac, A. Celic 

Fullerene Gear Design, Simulation, and Visualization 49 

Albert Globus 

Global Civil Aviation/Environmental Compatibility 

Civil Tiltrotor Noise Abatement Approaches 5^ 

William A. Decker, Rickey C. Simmons 

XV-15 Blade- Vortex Interaction Noise 52 

C. W. Acree, Megan 5. McCluer, Cahit Kitaplioglu 

Tiltrotor Aeroacoustic Model ^3 

Larry Young 

Predicting and Analyzing Rotorcraft Noise 54 

Roger C. Strawn, Rupak Biswas, Lenny Oliker 

Airframe Noise Measurements: Atmospheric Pressure , 55 

W. Clifton Home, ]ulie A. Hayes, Michael E. Watts, Paul H. Bent 

Airframe Noise Measurements: Pressures to 4.7 Atmospheres 55 

W. Clifton Home, Stephen M, jaeger, Mahendra joshi, James R. Underbrink 



Aeronautics Enterprise (continued) 

Revolutionary Technology Leaps/Innovative Technology and Tools 

X-36 Pioneers Advanced Aerodynamics and Flight Controls 

Rodney Bailey, Mark Sumich 
Closed-Loop Neural Control of Rotorcraft Vibration ^^ 

Sesi Kottapalli 
Coupled Navier-Stokes and Optimizer Analysis of a Transonic Wing ^^ 

Roxana M. Greenman, Samson Cheung, Eugene L Tu 

59 
Parallel Unstructured Mesh Adaption 

Rupak Biswas, Leonid Oliker, Roger Strawn 
Load Balancing Adaptive Unstructured Meshes.... 

Rupak Biswas, Leonid Oliker, Andrew Sohn 
Initial Release of the Field Encapsulation Library 



Steve Bryson 

MS-MP: Portal 
R. Van der Wijngaart, Maurice Yarrow 

allel Tools for Parallel and Distributed 
D. DiNucci, M. Frumkin, R. Hood, H. Jin, L Lopez, R. Papasin, C Scbulbach, ), Yan 



RANS-MP: Portable Parallel Navier-Stokes Solver ^^ 

Parallel Tools for Parallel and Distributed Computer Systems. ^^ 



Numerical Aerodynamic Simulation Parallel Benchmarks 2.2 ^^ 

W/7//am Saphir, R, Van der Wijngaart, Alex Woo, Maurice Yarrow 
Large-Scale Parallel Semiconductor Simulation 

Subhash Saini 

Revolutionary Technology Leaps/Supersonic Technology 

Nonlinear Aerodynamic Shape Optimization of High-Speed Research Configurations 66 

Susan Cliff, James Reuther, Ray Hicks 
Surface Operations Behavioral Evaluation Interim Testbed for High-Speed Research 68 

Mary K. Kaiser 

Revolutionary Technology Leaps/Access to Space 

Propulsion Checkout and Control System 

Ann Patterson-Hine 



VI 



C S T E S T S 



Space Science Enterprise 

Overview 71 

Progress in Exobiology 

Remote Analysis of Martian Surface Materials 74 

D. Blake, P. Sarrazin, D. Bish, D. Vaniman, S. Chipera, S. A, Collins, T. Elliott 

Stable Isotope Biogeocherrtistry of Hydrothermal Systems 76 

David ]. Des Marais 

Fossilization Processes in Thermal Springs 77 

jack Farmer, Sherry Cady, David J. Des Marais 

Molecular Biomarkers for Stromatolite-Building Cyanobacteria 78 

Linda i. jahnke, Roger E. Summons, Harold P. Klein 

Earth-Threatening Comets Leave Tell-Tale Dust Trails 80 

Peter fenniskens, David Morrison 

Capturing Cosmic Dust on Mir 81 

Kenji Nishioka, Ted Bunch, Mark Fonda, Glenn Carle, Sherwood Chang, James Ryder, Janet Borg 

Simple Peptides at Water-Membrane Interfaces 82 

Andrevi" Pohorille, Christophe Chi pot 

Silicon-Micromachined Gas Chromatography System 84 

Thomas Shen, James Suminto, Frank Yang, Daniel Kojiro, Glenn Carle 

Nitrogen Sources and Sinks on Early Earth 85 

David P. Summers 

Progress in Planetary Systems 

Planetary Rings 86 

]eff Cuzzi 

Planetesimal Formation in the Protopianetary Nebula 87 

Jeff Cuzzi 

PASCAL: A Mars Climate Network Mission 87 

Robert M, Haberle, David C. Catling, Steven C. Merrihew 

The Center for Star Formation 88 

D. Hollenbach, P. Cassen 

Energetic Trapped Particles near Jupiter 89 

John D. Mihalov 

Wavelet Software 90 

JeffScargle 

Time-Dependent Structures in Galaxies 90 

Bruce F. Smith, Richard A. Gerber, Richard H. Miller, Thomas Y. Steiman-Cameron 

Galileo Encounters Jupiter: Results from the Probe 91 

Richard E. Young 



vu 



Space Science Enterprise (continued) 

Progress in Planetary Systems (continued) 



Regolith Effects on Mars' Climate 92 

Aaron Zent 

The Nature of the Martian Oxidants 92 

Aaron Zent 

A Thernno-Acoustic Oxidant Sensor 93 

Aaron Zent 

Progress in Astrophysics 

Astrobiology in the Astrochemistry Laboratory 93 

Louis I. Allamandola, Scott Sandford, Max Bernstein, Robert Walker, Dave Deamer 

Spectrum Synthesis of Hot Water in Sunspots and Selected Cool Stars 94 

Duane F. Carbon, David Goorvitch 

A High-Altitude Site Survey for SOFIA 95 

Michael R. Haas, Leonhard Pfister 

Kepler Mission Educational and Public Outreach Software 96 

David Koch 

Mid-Infrared Studies of Diffuse Interstellar Material 97 

Thomas L. Roellig 

Infrared Observations of GO. 18-0.04 98 

Janet P. Simpson, Sean W. }. Colgan, Angela 5. Cotera, Edwin F. Erickson, Michael R. Haas, 
Mark Morris, Robert /-/. Rubin 

New 3.405-Micron Interstellar Emission from Organic Hydrocarbons 99 

Gregory C. Sloan, Jesse D. Bregman 

Progress in Space Technologies 

Assessment of the Cassini Command and Data Subsystem 101 

Edward A. Addy 

Automatic Telescope Project 101 

John Bresina 

Guide Star Tracker for Gravity Probe B Relativity Mission 103 

John H. Goebel 

Pulse-Tube Cryocooler Development 103 

Peter Kittel 

Automated Space System Experimental Testbed Project 105 

Christopher Kitts 

Amphion and Meta-Amphion 105 

Michael Lowry 



VUl 



C V TESTS 



Space Science Enterprise (continued) 

Progress in Space Technologies (continued) 

Focal-Plane Sensor Array Development for Astronomy in Space 1 06 

Mark E. McKelvey, Robert E. McMurray, jr., Craig R. McCreight 

Intelligent Execution for Autonomous Spacecraft ]08 

Barney Pell 

New Millennium Program Deep Space 1 Flight Software Program Management 108 

Scott Sawyer 

Human Exploration and Development of Space Enterprise 

Overview 1 1 -i 

Astronaut Health/Science 

Biochemical Markers of Bone Metabolism in a Rat Spaceflight Model 114 

Meena Navidi, Jeanne Wren, Sara Arnaud 

Cerebrovascular Responses Prior to Fainting 1-|5 

Kana Kuriyama, Toshiaki Ueno, Richard E. Ballard, Donald E, Watenpaugh, 
Suzanne M, Fortney, Alan R, Hargens 

Chronic Exposure to Hyper-G Suppresses Otolith-Spinal Reflex in the Rat 1 1 6 

Nancy G. Daunton, Merylee Corcoran, Robert A. Fox, Li-Chun Wu 

''Dual Adaptation'' to Space-Related Sensory Rearrangements 118 

Roberts. Welch 

Technology Applications to Human Health 

Virtual Environment Surgery Workbench 1 19 

Muriel D. Ross 

Noninvasive Estimation of Pulsatile Intracranial Pressure Using Ultrasound 1 20 

Toshiaki Ueno, Richard E. Ballard, John H. Cantrell, William T. Yost, Alan R. Hargens 

MRI-Compatible Spinal Compression Harness 122 

Richard E. Ballard, Donald E. Watenpaugh, Iwane Mitsui, Klaus P. Fechner, 
Douglas 5. Schwandt, Alan R. Hargens 

Near-Infrared Spectroscopy to Monitor Forearm Muscle Oxygenation 1 23 

Gita Murthy, Alan R. Hargens 

BIONA 1 — Blood Flow Ion Analyzer -I25 

John W. Hines, Christopher J. Somps 

Intelligent Controller for Neurosurgery -|25 

Robert Mah 



IX 



Human Exploration and Development of Space Enterprise (continued) 

Technology Applications to Human Health (continued) 

Center for Health Applications of Aerospace Related Technologies (CHAART) 1 28 

Byron Wood, Louisj Beck, Sheri Dister, Brad Lobitz 
TelemedicineSpacebridge to Russia 

Steve N. Kyramarios 

Progress in Improving Space Travel 

Advanced Life Support/Human Exploration and Development of Space Enterprise Activities 1 29 



Dick Lamparter, Mark Kliss 

Hardware — Stepping Stone 

Bonnie P. Dalton, fames Connolly, Cary jahns, Paul Savage 



1 ^0 
Mir Hardware— Stepping Stone to Station 

/ jahns, Paul Savage 

131 



Wireless Network Experiment for Space Shuttle/Mir 

Richard Alena 

1 32 
Engine Diagnostic Filter System • 

Tarang Patel 

Formal Lightweight Approaches to Validation of Requirements Specifications 1 33 

5feve Easterbrook 

Astronaut Health/Countermeasures 

Autogenic-Feedback Training as a Potential Treatment for Postflight Orthostatic Intolerance 1 3S 

Patricia S. Cowings, William B. Toscano 

1 ^H 
Exercise for Long-Duration Spaceflight 

Donald E. Watenpaugh, Richard E. Ballard, Karen j. Hutchinson, laqueline M, William, 

Andrew C. ErtI, Suzanne M. Fortney, Lakshi Putcha, Wanda I, Boda, Stuart M. C Lee, 

Alan R. Hargens 

1 -JQ 

Keiser SX-1 Variable Resistance Exercise Device 



Jennifer Pedley, Anthony Artino, Richard Ballard, Alan R. Hargens 

lydration at Airline Cabin Altitude 

John E. Greenleaf, Peter A. Farrell, Helmut Hinghofer-Szalkay 



Dehydration at Airline Cabin Altitude ^"^^ 



C \ i I \ J s 



Mission to Planet Earth Enterprise 

Overview 145 

Ecosystem Science and Technology 

AIRDAS — Use of Remote Sensing for Disaster Assessment and Management 147 

James Brass, Vincent Ambrosia, Robert Slye 

Bay Area Digital Georesource 147 

Edwin Sheffner, Sheri Dister, Don Sullivan 

Brazil/United States Environmental Monitoring and Global Change Program 148 

James Brass, Vincent Ambrosia 

Carnegie/Ames/Stanford Approach Model 149 

Christopher Potter 

Digital Array Scanning Interferometer 149 

Steve Dunagan, Philip Hammer 

Effect of Landuse on Regional Estimates of Coniferous Forest Water and Carbon Budgets 151 

Joseph Coughlan, Jennifer Dungan 

Landsat Program 152 

Edwin Sheffner 

Leaf Modeling 1 52 

Lee F. Johnson, Chris Hiavka, Philip D. Hammer, David /.. Peterson 

Mapping Northern Ecosystems: Applications for Circumpolar Methane Exchange 1 53 

Vern Vanderbilt, Cuillaume Perry, Joel Stearn 

Modern Ecosystems Research: Effects of Increased UV-B Radiation 1 53 

Hector L. D'Antoni, J. W. Skiles 

Optimizing an Ecosystem Model for Use on Parallel/Distributed Processors 1 54 

;. W. Skiles, Cathy Schulbach 

Paleoenvironmental Research 155 

Hector L, D'Antoni 

Scientists' Intelligent Graphical Modeling Assistant 155 

Jennifer Dungan 

Atmospheric Chemistry 

Airborne Natural Radionuclide Measurements in the Development and Validation of 

Global Three-Dimensional Models 1 56 

Mark Kritz, Stefan Rosner, Robert Chatfield, Leonard Pfister 

Reactive Nitrogen Data from the Upper Troposphere and Lower Stratosphere 1 56 

Hanwant B. Singh, Alakh Thakur, Peter Mariani 

Airborne Autotracking Sunphotometry 1 57 

Philip B. Russell, John M. Livingston, James Hanratty, Damon Ried, Jill Bauman 



XI 



Mission to Planet Earth Enterprise (continued) 

Atmospheric Chemistry (continued) 

Analysis of Stratosphere/Troposphere Exchange 157 

Leonhard Pfister, Henry Selkirk 

Use of Argus in Atmospheric Studies 158 

Max Loewenstein 

Airborne Tunable Laser Absorption Spectrometer 158 

Max Loewenstein, James R. Podolske 

Convectively Generated Gravity Waves 159 

Leonhard Pfister 

The ER-2 and DC-8 Meteorological Measurement Systems 1 59 

K. Roland Chan, T. Paul Bui, Antonio A, Trias, Stuart W. Bowen, Jonathan Dean-Day 

Environmental Research Aircraft and Sensor Technology 160 

Steve Wegener 
Global Emissions Inventories for Radon and the Cosmogenic Radionuclides 161 

Mark Kritz 
The Great African Plume: Tropical Carbon Monoxide and Ozone Simulation 1 62 

Robert B. Chatfield 
Instrument for Tropospheric Nitrogen Studies 163 

James R. Podolske 
Reactive Nitrogen and Oxygenated Hydrocarbon Measurements during the Pacific 
Exploratory Mission 163 

Hanwant B. Singh, W, Viezee, R, Chatfield, Y, Chen, D, Herlth, R. Kolye 

Subsonic Aircraft: Contrail and Cloud Effects Special Study 163 

Owen B, Toon, Steve Hipskind, Duane Allen, Paul Bui, Roland Chan, Mike Craig, Guy Ferry, 
Steve Gaines, Warren Gore, Eric Jensen, Joe Jordan, Stefan Kinne, Bill McKie, Peter Pilewskie, 
Rudi Pueschel, Tony Strawa, Annette Walker 

Stratospheric Tracers of Atmospheric Transport 164 

Stephen Hipskind, Michael Craig 
Tropospheric Aerosol Radiative Forcing Observational Experiment 165 

Philip B. Russell, John M. Livingston, Wendy Whiting 

Atmospheric Physics 

Fine Particle Emissions by Aircraft 166 

Rudolf F. Pueschel, Guy V. Ferry, Anthony W. Strawa, Duane Allen 

FIRE Phase III 166 

Peter Pilewskie, Warren Gore 

Laboratory Spectroscopy of Carbon Dioxide in Support of Planetary Atmospheres Research 1 67 

Lawrence P. Giver, Charles Chackerian, jr. 



Xll 



C S T E S T S 



Mission to Planet Earth Enterprise (continued) 

Atmospheric Physics (continued) 

Near-Infrared Remote Sensing of Cloud Liquid Water 

Peter Pi lewskie, Warren Gore 

Quantitative Infrared Spectroscopy of Minor Constituents of the Earth's Atmosphere i Aft 

Charles Chackerian, jr., Lawrence P. Giver 

Stratospheric Transport 
Rudolf F. Pueschel, 

SUCCESS Irradiance Measurements 



Rudolf F. Pueschel Guy V. Ferry, Anthony W. Strawa, Duane Allen 



Peter Pilewskie, Warren Gore ^ ^^ 

Appendix 

Color Plates (1-22).. 

172 



XIU 



ansportation Technology Enterprise 






A E R \ A I r I c s A\ D Space T r a s s p o r r a t i o \ 

T E C H S I C y E \ r ER P R I S E 



Overview 



NASA's mission for the Aeronau- 
tics and Space Transportation Tech- 
nology (ASTT) Enterprise is to pioneer 
the identification, verification, 
transfer, application, and commercial- 
ization of high-payoff aeronautics and 
space transf)ortation technologies. 
Ames' researchers and technologists 
support this mission by seeking 
enabling revolutionary technological 
advances that will provide air and 
space travel for anyone, anytime, 
anyw^here more safely and more 
affordably, and v^ith less effect on the 
environment and with improved 
business opportunities and global 
security. 

This pursuit of revolutionary 
technologies addresses bold, mid- 
term and long-term goals of the 
ASTT Enterprise to enable dramatic 
improvements in aviation and space 
transportation. This reflects national 
priorities as outlined by the National 
Science and Technology Council, 
These goals are grouped into three 
areas, or as they are called, three 
pillars: "Global Civil Aviation," 
"Revolutionary Technology Leaps/' 
and "Access to Space." The following 
sections outline these goals and 
Ames' FY96 accomplishments toward 
achieving them. 

Pillar 1: Global Civil Aviation 

Today, with over 1 1,000 air- 
planes in commercial service world- 
wide, the United States faces strong 
international competition in this vital 
area whose products are the largest 
positive industrial contributor to the 
U.S. balance of trade. Projects linked 
to world economic growth suggest 
that air travel demand will triple over 
the next 20 years. Therefore, to 
preserve our Nation's economic 
health and the welfare of the traveling 
public, NASA must provide high-risk 
technology advances that will 
contribute to safer, more affordable, 
more environmentally compatible air 
travel. 



The highlighted FY96 accom- 
plishments address the following 
goals of the Global Civil Aviation 
pillar: 

1 . Safety goal: Reduce the aircraft 
accident rate by a factor of 5 
within 10 years, and by a factor 
of 10 within 20 years. 

2. Affordability goals: (a) While 
maintaining safety, triple the 
aviation system throughput, in 
all weather conditions, within 
10 years and (b) reduce the cost 
of air travel by 25% within 

10 years, and by 507<) within 
20 years. 

3. Environmental compatibility goal: 
Reduce the perceived noise levels 
of future aircraft by a factor of 2 
(from those of today's subsonic 
aircraft) within 10 years, and by a 
factor of 4 within 20 years. 

Research and development 
conducted by the ASTT Enterprise has 
been structured to be led by specified 
NASA research centers according to 
the primary roles and missions that 
have been assigned to each center. 
Ames is the lead center for the High 
Performance Computing and Com- 
munications (HPCC) Program and for 
fhe research and technology (R&T) 
base programs in aviation operations 
systems, information technology, and 
rotorcraft. In addition, Ames leads the 
Enterprise core competencies in the 
areas of human factors, air-traffic 
management, information system 
technologies, and rotorcraft R&T. 

Ames' significant involvement 
in the safety goal includes human 
factors, information technology, and 
condition-based maintenance. Ames 
made significant contributions toward 
addressing this goal, contributions 
that led to advances in aviation 
performance measuring, aviation 
safety reporting, crew activity track- 
ing, air-traffic complexity, aircraft- 
separation risk modeling, air and 
ground integration, and human error 
in aircraft maintenance. 

Aidines and businesses lose 
billions of dollars annually as a result 



of delays and lost productivity owing 
to weather and congestion in the 
airspace system. Under affordability 
goal (a) (above), Ames' major effort 
in research and development of air- 
traffic management automation 
constitutes virtually all of NASA's 
work in that area. Accomplishments 
include articles on controller perfor- 
mance under simulated free-flight 
conditions, techniques for low- 
visibility taxi and ground-collision 
avoidance, and conflict-prediction 
algorithms. 

Reducing the costs of aircraft 
operation and maintenance is a major 
challenge. NASA's test facilities and 
core expertise in materials, structures, 
aerodynamics, propulsion, analytical 
methods, and computational tools are 
key elements in helping to revolution- 
ize aircraft design and manufacturing. 
NASA's research efforts are focused 
on Innovative design techniques and 
structural concepts. Ames' contribu- 
tions to affordability goal (b) come 
from rotorcraft, national tasking 
facilities, and research and develop- 
ment of computational tools. Several 
Ames projects contributed to rotor- 
craft research and technology 
activities, which include rotor aero- 
dynamics, testing of new concepts, 
stall control, in-flight dynamic stall 
research, and vortex flows. Examples 
of accomplishments that enhance 
computational tools are vortex-core 
detection and tracking of particles in 
time-varying flows. Improvements in 
the use of facilities are highlighted in 
the following pages by articles on 
planar Doppler velocimetry, pressure- 
sensitive paint and photogrammetry, 
and visualization of experimental 
data. A high-risk, long-term research 
effort directed to the manufacture of 
fullerene nanotechnology gears is also 
described. 

For research related to the 
environmental compatibility goal, 
Ames has significant aeroacoustic 
research and testing capabilities, the 
most recent advance in those capa- 
bilities being the Aeroacoustic 



Modification Project at Ames' 
National Fuil-Scale Aerodynamics 
Complex (NFAC), which is currently 
being completed. In this report, 
accomplishments related to helicopter 
noise and airframe noise are 
presented. 

Pillar 2: Revolutionary Technology 
Leaps 

NASA's charter Is to explore high- 
risk technology areas that can revolu- 
tionize air travel and create new 
markets for U.S. industry. The tech- 
nology challenges for NASA Include 
accelerating the application of 
technology advances, eliminating 
the barriers to affordable supersonic 
travel, and expanding general 
aviation, 

The highlighted FY96 accom- 
plishments address the following two 
goals of the Revolutionary Technol- 
ogy Leaps pillar: 

1 . Provide next-generation design 
tools and experimental aircraft to 
increase design confidence, and 
cut the aircraft development 
cycle time in half. 

2. Reduce the travel time to the Far 
East and Europe by 50% within 
20 years, and do so at today's 
subsonic ticket prices. 

The next-generation design tools 
and experimental aircraft goal will 
dramatically affect the way in which 
business is conducted. Its effect will 
be felt across the three pillars, contrib- 
uting to every technology goal. Ames 
has significant work in Integrated 
design systems and the X-36 aircraft. 
Research is done at Ames in informa- 
tion technology to elevate the power 
of computing tools through fuzzy 
logic, neural networks, and artificial 
intelligence. These tools will integrate 
multidisclplinary product develop- 
ment activities to dramatically cut 
design cycle times. Examples of 
accomplishments include neural 
control of rotorcraft vibration, cou- 
pling of flow-field computation tools 
with design optimization tools, and 
parallel computational tools and 



computations. Experimental aircraft 
are also invaluable toots for exploring 
new ideas. An example Is provided 
with the use of advanced aero- 
dynamics and flight controls on the 
X-36 aircraft. 

Under the high-speed transport 
goal, Ames has diverse activities in 
wind-tunnel testing and simulation, 
external visibility, sonic boom 
minimization, and wing aerodynamic 
optimization. Articles are presented In 
this report to highlight these activities. 

Pillar 3: Access to Space 

In coming decades, NASA 
envisions the space frontier as a busy 
crossroads of U.S.-led International 
science, research, commerce, and 
exploration. Experience with this vast 
resource has already yielded new 
treasures of scientific knowledge, life- 
enhancing applications for use on 
Earth, and fantastic celestial discover- 
ies. The potential for the future seems 
almost limitless. 

Ames addresses the following 
goal of the third pillar. Access to 
Space: 

1 . Reduce the payload cost to 
low-Earth orbit by an order of 
magnitude, from $10,000 to 
$1 ,000 per pound, within 
10 years. 

As NASA's lead center for 
thermal protection systems (TPSs) 
technology, Ames is charged with 
developing new thermal protection 
systems that will enable vehicles 
of the future to be built more eco- 
nomically and that will enable 
existing ones to be upgraded at 
reduced cost. Ames maintains one of 
the world's premier arc-jet complexes 
for providing realistic simulations of 
entry environments. These simulations 
are essential for technology develop- 
ment, system validation, and system 
qualifications. Ames supports the U.S. 
aerospace community in developing 
TPSs that will be necessary for the 
nation's future space vehicles. 



A E R S A I r I ( S A S D S P A ( E T R A \ S P R I A T { A 
T F C H \ I G y E \ T f R P R I S E 



Aviation Performance 
Measuring System 

Irving C. Statler 

Flight Operations Quality 
Assurance (FOQA) programs using 
tlight-rec:orded data have been 
providing c riticai safety information 
to non-U. S. airlines for over tw^o 
decades. Although the benefits of 
these programs have been demon- 
strated, the U.S. air carriers have 
found them impractical to imple- 
ment because of the extensive 
labor required to process the great 
amounts of data that would typically 
be generated. A collaborative effort 
was initiated in August 1993 
between the Federal Aviation 
Administration (FAA) and NASA to 
establish and demonstrate the 
feasibility of developing a set of 
tools that would allow very large 
quantities of flight data to be pro- 



cessed automatically in order to 
address questions relating to opera- 
tional performance and safety. 

The Aviation Performance 
Measuring System (ARMS) is provid- 
ing technical tools to facilitate the 
large-scale implementation of flight- 
data analyses at both the air-carrier 
and the national-airspace levels. 
ARMS enhances the existing 
Commercial Off The Shelf (COTS) 
cafjabilities of FOQA, including the 
caf)ability of analyzing all the data 
collected, in addition to merely 
identifying ''exceedances" or 
"special events" (the figure shows 
the APMS functions). 

Phase 1 of the APMS effort 
ended in December 1995. Phase 2 
focuses on three tasks: (1) selection 
and implementation of an APMS 
client-server database architecture; 
(2) development of a knowledge- 
based system for verifying and 
diagnosing "special events" flagged 
by COTS packages; cincl (3) construc- 
tion of a friendly user-interface in 



e — 














Screening lor 
special events 

A 


Flight data 
inputs 


Statistical 
^ analysis 


h 


/\ 


^ ^ 


i 


». 


W! 19M 1!W: IMB 199* JOCM 


Database 
exploration 


c ^ 




Database 

I 








\ 






Database 
linkage 


Flight 
animation 

















Visual Basic. The client-server 
database architecture and method 
for the initial build have been 
selected. Teledyne Control's 
FLIDRAS software has been acquired 
to reixd Flight Data Recorder raw 
data parameters, to convert the data 
into engineering units, and to 
perform data dumps into the data- 
base management system. As the 
flight database is built, the baselines 
ot various routine operations will be 
established. Knowledge-based tools, 
currently under development by the 
APMS team, will, in later iterations 
when an adequate database is 
available, provide statistic al trending 
and predictive capabilities. The first 
prototype system was delivere^d in 
July 1996 to the first airline partner. 
The prototype system is now opera- 
tional and currently being used to 
process airline flight data. 

With the industry becoming 
aware of the cat^abilities of the 
APMS suite of tools, the team is 
being approached by other airlines 
that had not been solicited to 
participate in the research project. 
So far, one additional major airline 
and one major cargo airline have 
asked to participate in the program. 
User-needs studies have been 
completed with four U.S. airlines 
and a fifth has been started. Agree- 
ments have been signed with two 
of these to cooperatively develop 
customized suites of APMS tools. 
The APMS effort continues to work 
with collaborators to improve system 
design. 

Point of Contact: I. Statler 
(650) 604-6655 
istatler@mail.arc.nasa.gov 



Fig. 1 . APMS functions. 



(i i. M A L {] M I I. A \ [ A T I (1 N / Sd/i'/V 



Aviation Safety Reporting 
System 

Linda J. Connell 

The Aviation Safety Reporting 
System (ASRS) solicits, processes, 
and analyzes aviation safety incident 
reports from pilots, air-traffic control- 
lers and others and uses the data it 
collects to further aviation safety. 
The ASRS (1 ) codifies the reports it 
receives (31 ,096 in FY96) and inserts 
them into a computer database; 
(2) issues alerting messages on 
pressing safety problems described 
by incoming reports; (3) provides 
data retrieval services (search 
requests) for aviation safety research- 
ers and others; (4) publishes a 
monthly safety bulletin, CALLBACK, 
and a periodic safety journal, 
DIRECTLINE: (5) performs Quick 
Response analytic efforts for the 
Federal Aviation Administration 
(FAA), the National Transportation 
Safety Board (NTSB), and other 
governmental entities; and (6) does 
applied research on aviation opera- 
tional problems, especially those 
involving human performance (see 
the figure). The ASRS is heavily 
reliant on information technology, 
and a portion of program resources 
is devoted to maintaining and 
upgrading that technology. 

The following list is a summary 
of some of ASRS's FY96 activities, in 
FY96, the ASRS program: 
• Initiated an ASRS Internet site in 
November 1995. Through its 
Web site, ASRS offers its publi- 
cations, program information, 
and downloadable (using an 
Adobe Acrobat reader) versions 
of the NASA Incident Reporting 
Forms. 



6 

Quick responses 
to FAA & NTSB 




Database 
CD ROM 




^^THMli Pir^^ine^ 



Monthly safety 
newsletter 



Quarterly safety 
bulletin 



Research 



Fig. 1 . ASRS products. 



• Accomplished a Multi-Engine 
Turbojet Uncommanded Upsets 
Structured Callback Analysis for 
the NTSB. The analysis was 
subsequently adopted into the 
USAir B-737 accident investiga- 
tion report. 

• Completed a GPS Safety Impact 
Analysis for the FAA Associate 
Administrator of Research and 
Acquisitions. 

• Accomplished a Quick Response 
analysis of Runway Transgres- 
sion Data for the FAA Office of 
System Safety. These data were 
subsequently presented to the 
FAA deputy administrator. 

• Completed Quick Response 
No. 289 entitled "Near Mid-Air 
Collision Data Analysis" for the 
FAA Office of System Safety. 



Accomplished the second 
installment of the Wake Turbu- 
lence Structured Callback 
Project Reports; it included the 
analysis of 51 wake turbulence 
incidents and was submitted to 
the FAA. 

Produced search requests for the 
NTSB and the FAA on DC-9 
cabin/cockpit smoke and aircraft 
equipment problems in support 
of the investigation of the DC-9 
accident near Miami, Florida, on 
May 11, 1996. 
Produced a Quick Response 
analysis of Part 1 35 Aircraft 
Incidents for the Australian 
Bureau of Air Safety Investiga- 
tion. The data are being used to 
develop a proactive aviation 
safety hazard monitoring system. 



(i !. H A I C I V I L A V 1 A T 1 N / Safely 



A t R \ .\ ! in s 



\ I) S p I ( t: T R A \ V p R r .\ T I s 
T I ( u A i) I (r 1" E \ r t: R p R I s t 



• Transmitted 30 runway- 
incursion reports at Cleveland- 
Hopkins Airport, Ohio, to the 
FAA Office of System Safety and 
the Air Line Pilots Association. 
As a result of ALPA and FAA 
follow-up actions, revised 
ground-control procedures were 
implemented for departures 
using runways 23L/23R. 

• Transmitted 24 reports to FAA 
and ALPA involving altitude 
deviations at FUELR Intersection 
on the CIVET ONE ARRIVAL to 
Los Angeles International 
Airport. Subsequent investiga- 
tion by the FAA determined that 
the recent installation of a new 
ILS had changed the glidepath 
angle at FUELR. 

• Published one issue of 
DIRECTLINE in FY96. Its 
research articles addressed 
aircraft call-sign confusion and 
ramp safety. 

• Developed a trial version of the 
ASRS Intranet communications 
system. It runs on an NT Server 
and is available to the ASRS 
staff. The decision was made in 
January 1996 to develof) this 
internal Intranet communica- 
tions system based on existing 
Internet applications in order to 
lower costs and increase effi- 
ciency internally at the ASRS. 

• Commenced a collaborative 
research project between 
NASA and its French counter- 
part, ONERA, to evaluate an 
ONERA human factors coding 
scheme and taxonomy for 
possible application to ASRS 
incident data. 



• Completed a conceptual and 
editorial review for a second 
NASA project, a study on the 
use of modes in human-machine 
interactions. This study, w^hich 
was completed, made use of 
ASRS data. 

Point of Contact: L. Connell 
(650) 604-6654 
lconnell@mail.arc.nasa.gov 



Crew Activity Tracking 
System 

Todd J. Callantine 

FHuman operators supervising 
advanced automation systems can 
have difficulties that have the 
potential to compromise safety. For 
example, a recent study found that 
over 44'X) of flight deck problems 
cited in a large body of incident and 
accident ref)orts were related to 
automation. Autoflight system 
modes, in particular, are often 
implicated because they can cause 



unexpected behavior. One remedy 
is to develop technology that can 
detect potential operator errors and 
provide the operator with context- 
sensitive advice and reminders. To 
be effective, such technology must 
incorporate knowledge about 
operator-automation interaction in 
context; for flight deck application, 
this includes knowledge about mode 
management. 

This research addresses an 
enabling technology called activity 
tracking, a way of inferring intent. As 
the name implies, the Crew Activity 
Tracking System (CATS) is an 
architecture that implements a 
method for activity tracking (see the 
first figure). CATS represents knowl- 
edge about the operator's tasks using 
an explicit, task-analytic model 
based on the goal-based decomposi- 
tion of tasks into subgoals and 
primitive tasks. CATS uses the model 
to t)redict operator intentions and to 
interpret subsequent operator 
actions. The predictions and inter- 
pretations are designed to supply the 
knowledge required for intelligent 
aiding and training systems. 




CATS 



Human controlled 
operators system 



State ^ 
information "^ 



Constraints ^^ 
on operation 



^Context^ 
specifiers 



Crew activity model 



IL 



OFM- 
ACM 



Predictions 

*- — 

I 



Crew 
actions 



Action 
manager 



Interpretations 



-[to aid or training system] - 



Fig, 1 . The Crew Activity Tracking System in context. 



(i 1. H A 



(^ 1 \ 1 L A \ I \ T [ N / Safety 























access-LEGS-pg 


— 


push-LEGS-key 
















buJId-ctas-adp-ref-wpt 


— 


put-ref-wpt-scr-pad 
















enter-ctas-adp-ref-wpt 


— 


put-ref-wpt-LEGS-pg 






















program-ctas-adp 






build-ctas-adp-ref-dist 


— 


put-ref-dist-scr-pad 




























enter-ctas-adp-ref-dist 


— 


put-ref-dist-LEGS-pg 
















verify-ctas-adp 


— 


SOP-talk 
















execute-ctas-adp 


— 


push-EXEC-key 





















Fig, 2. A simplified task decomposition from the enhanced CATS model. 



CATS was implemented to track 
the activities of Boeing 757 pilots 
using autopilot flight modes under 
normal operating conditions. An 
experimental evaluation was con- 
ducted to assess the effectiveness of 
the CATS method. Ten type-rated 
line pilots from a major airline 
participated in the study. Each flew 
five experimental scenarios on a 
real-time part-task simulation of the 
Boeing 757. Of the 2,089 pilot 
actions detected in the study, CATS 
correctly interpreted 81%. A subse- 
quent analysis identified adjustments 
to the CATS model and processing 
scheme that would allow CATS to 
correctly interpret 947o of pilot 
actions. 

Current research seeks to extend 
the capabilities of CATS and to 
explore new applications. Specifi- 
cally, the objective is to enable 
CATS to predict and interpret the 
actions of two-person crews by using 
complex functions of the Flight 
Management Computer Control and 
Display Unit (FMC CDU), and data- 
link communications. These activi- 
ties are critical for procedures that 



are being developed for use with 
new air-traffic control automation. 

The second figure shows a 
simplified task decomposition from 
the enhanced CATS model, from 
which most cognitive, verbal, and 
perceptual activities are omitted. The 
task shown supports a new descent 
procedure that uses an assigned 
descent point C'adp"), in addition to 
the top-of-descent point computed 
by the FMC. When the crew has 
been cleared for the procedure, 
CATS uses this portion of the model 
to predict and interpret the CDU 
programming activities required to 
perform the task. For example, CATS 
first predicts which pilot is respon- 
sible for programming the assigned 
descent point to go to the CDU 
LEGS page. CATS next predicts that 
a reference way point will be built in 
the CDU scratchpad, then line- 
selected to the appropriate place on 
the LEGS page, and so on. As these 
actions are performed, CATS checks 
the entered values for accuracy and 
then either interprets the actions to 
support the task of entering the 



assigned descent point or signals a 
potential error. 

A Boeing 747-400 simulator 
study to investigate pilot perfor- 
mance on a new descent procedure 
was conducted at Ames Research 
Center. The data show that although 
pilots performed the new procedure 
effectively in most cases, there were 
still a number of compliance viola- 
tions. These data are used to investi- 
gate a new application of CATS: 
tracking crew activities to automati- 
cally identify departures from the 
procedure and capture the context 
in which these departures occurred. 
With this information, the procedure 
may be refined to address problems, 
and thereby improve compliance. 

Point of Contact: T. Callantine 
(650) 604-2631 
tcallantme@mail.arc.nasa.gov 



Study of Line Oriented 
Flight Training 

Key R. Dismukes 

Airlines train pilots to work 
together as safe, efficient crews by 
using a realistic full-mission flight 
simulation approach called Line 
Oriented Flight Training (LOFT). In 
this annual recurrent training, crews 
encounter challenging situations that 
they must manage by coordinating 
their efforts, exercising good judg- 
ment and decision-making, and 
drawing upon all available 
resources. After the LOFT, the 
instructor leads the crew in a 
debriefing in which they are 
expected to analyze what happened, 
evaluate their own performance, 



C [ \" I 1. A V i A T ! N / Safely 



A t R a \ \ i in \ 



1 \ I) S F \ ( t: T R A S S P R T i T { S 
T t { If \ L G ] E \ r tR P R I S E 



and identity ways to improve 
performance. Because the LOFT is a 
very busy, intense experience, how 
much the crews learn from the LOFT 
and lake back to line operations 
hinges on the effectiveness of the 
debriefing. 

Both the airlines and the FAA 
espouse the idea that LOFT instruc- 
tors should refrain from lecturing the 
crews in the traditional ''teacher-tell" 
manner. Instead, instructors are 
encouraged to "facilitate" self- 
analysis by the crew so that they will 
learn more deeply. However, there 
has been little study of how to 
facilitate this self-analysis; as a 
consequence, the airlines have not 
had good data-based techniques for 
training their instructors in debriefing 
procedures. 

Human factors scientists at 
Ames recently completed a study, 
in collaboration with major U.S. 
airlines, to evaluate LOFT debrief- 
ings and to provide guidelines for 
training instructors. The study 
evaluated the effectiveness of LOFT 
debriefings at a cross section of 
airlines, analyzed facilitation 
techniques, and identified common 
errors. The study demonstrated that 
instructors who are effective in 
facilitating self-analysis substantially 
increase the depth of crew participa- 
tion and self-analysis; however, 
individual instructors differed greatly 
in their effectiveness as facilitators. 

In conjunction with the techni- 
cal report of the study, the research 
team prepared a detailed manual for 
use in training instructors to facilitate 
debriefings. This manual explains 
the concepts of facilitation, describes 
effective techniques, explains how 



and when to use these techniques, 
and shows how to engage crews that 
do not initially respond. Immediately 
after receiving the study materials, 
several major airlines reprinted the 
manual for the use of their instruc- 
tors. Other airlines have overhauled 
their training of instructors to 
incorporate the findings from the 
study; they later reported significant 
benefits to their crew training 
programs. 

Point of Contact: K. Dismukes 

(650)604-0150 

kdismukes@mail.arc.nasd.gov 



Measuring Air Traffic 
Complexity 

Irene V. Laudeman, Connie Brasil, 
Robert Branstrom 

An important goal in advanced 
air-traffic operations is that of 
proviciing more system flexibility to 
the user. The idea of sharing the 
responsibility for aircraft separation, 
in which pilots and air-traffic 
controllers work together to ensure 
separation, is designed to provide 
such flexibility. Currently, under 
instrument flight rules the air-traffic 
controller is tasked with ensuring the 
proper separation of aircraft. Under 
shared separation-responsibility 
procedures, pilots in appropriately 
equipped aircraft would assume 
some responsibility for the manage- 
ment of aircraft separation. The 
proportion of responsibility assigned 
to pilots and air-traffic controllers 



would be a function of the complex- 
ity of the airspace. 

Shifts in operational procedures 
will be specified as a function of 
airspace complexity, making the 
development of separation proce- 
dures dependent on the develop- 
ment of a complexity metric. The 
complexity of air-traffic patterns, and 
hence the workload related to the 
management of air traffic, is known 
to include more than a simple count 
of aircraft contained in a volume of 
airspace. However, the traffic 
complexity factors beyond that of 
aircraft number have not been 
clearly identified, nor has any 
relationship among the factors been 
established. 

A human-in-the-loop simulation 
study was conducted to evaluate 
three modes of separation resf)onsi- 
bility and to provide data for use in 
developing a complexity metric. Ten 
groups of four currently qualified air- 
traffic controllers participated in nine 
simulation scenarios. In each group 
one participant acted as the control- 
ler in a sector of airspace and three 
participants acted as pilots. 

Aircraft data in the form of 
latitude, longitude, heading, altitude, 
and speed were collected for all of 
the aircraft in each of the nine 
30-minute scenarios. The complexity 
of the simulated airspace was 
computed from the aircraft data at 
every 2 minutes of the 3()-minute 
scenarios. Complexity functions 
were computed as a sum of traffic 
factors that included aircraft count, 
aircraft converging at the same 
altitude, and aircraft changing 
heading, speed, or altitude. 



(J 1, H A 1. (] 1 \ I [. A \ I A i i N / Safety 



Shift in separation responsibility 



140 



(0 

o 80 

O 



Scenario 1 sliift 
Scenario 1 shared (3) 
Scenario 2 shift 
Scenario 2 shared (7) 




12 14 16 
Time (min) 



26 



Fig. 1 . Air-traffic complexity functions and points at which separation 
responsibility was shifted for two types of traffic scenarios. 



The complexity functions 
predicted an operational shift related 
to the sharing of separation responsi- 
bility (see the figure) indicating that 
such functions might be useful in 
specifying shared separation respon- 
sibility procedures. 

Point of Contact: I. Laudeman 

(650)604-0018 

laudeman@eos.arc.nasa.gov 



Operational Interventions 
to Human Error in Aircraft 
Maintenance 

Barbara G. Kanki, Vicki Dulchinos 

A significant proportion of 
aviation accidents and incidents are 
known to be tied to human error. 
However, research of flight opera- 
tional errors has shown that 
so-called ''pilot error" often involves 
a variety of human factors issues 
and not a simple lack of individual 
technical skills. In aircraft mainte- 
nance operations, there is similar 
concern that maintenance errors 
which may lead to incidents and 
accidents are related to a large 
variety of human factors problems. 



Although industry initiatives involv- 
ing human factors training in 
maintenance have become increas- 
ingly accepted as one type of 
maintenance error intervention, 
there remains a dual challenge: 

(1) to develop human factors inter- 
ventions that are directly supported 
by reliable human error data, and 

(2) to integrate human factors 
concepts into the procedures and 
practices of everyday technical tasks. 

These challenges are being 
addressed at Ames Research Center. 
First, industry-wide incidents 
reported to the Aviation Safety 
Reporting System are being analyzed 
in order to identify and characterize 
high priority problem areas. When 
analyzed for contributing factors, at 
least 507o of the incidents involved 
more than one technician — other 
maintenance personnel, flight crew 
members, and airport personnel. 

Consistent with these findings, 
research is being conducted that 
focuses on human factors interven- 
tions related to practices and 
procedures; namely, structured 
on-the-job training and procedure 
re-design, in both areas, particular 
attention is being centered on areas 
in which maintenance tasks require 
coordination both within and 
between maintenance teams. 

In the area of structured on-the- 
job training, field applications of 
the Task Analytic Training System — 
a performance-based system that 
involves full workforce participation 
in its design, development, and 
implementation— have been 
conducted. 



(I i. H A [. (1 1 V M, A V i A T ION / Safely 



Aero s a i r i c s a v d Space T r a \ s p o r r a r ! o s 

T E C H \ I G y E \ T E R P R 1 S E 





A. 5°/ 


'o 




X 


jfl 


p 


B. 8% 

\ C. 12% 




^^^^B^^^^^^S"^***^ 








^^R^^^^a 




— ^.^ D. 4% 




\^'"^^^^^^fcM^^^- '"^^^^^^1-^1 




7 


'■''""'' ""^^^^S 


> 


/ 




"*««^^re«i 


^^ 


E. 19% 










■ 


A. Situation awareness/Time task 


management (n=38) 


■ 


B. Planning/Communication-new resources (n=62) 


1 1 


C. Planning/Communicatior 


\ of task sequence {n=85) 


1 1 


D. Clarification/Formatting - 


Simplified English (n=28) 


u 


E. Clarification/Formatting consistency (n=139) 


11 


F. Process improvement (n: 


=243) 




■ 


X. No functional change (n= 


140) 





¥\^. 1. Distribution of reason codes in the revised procedure 
(N = 735 procedural! steps). 



Human factors principles may 
also be incorporated into the 
maintenance procedures themselves. 
When Boeing modified its B-737 
CFM56-7 engine-change procedure, 
the efficiency of the engine-change 
process was significantly improved. 
A ]4% increase in efficiency, 
attributed to changes in the mainte- 
nance manual alone, presented an 
opportunity to conduct a systematic 
comparison between original and 
revised procedures. The goal was to 
analyze the specific modifications 
that led to the improvement in 
efficiency and to identify the contri- 
bution of human factors. Some 



results of this item-by-item compari- 
son are found in the summary of 
functional changes shown in the 
figure. 

Finally, a collaborative relation- 
ship has been established with the 
human factors team at Kennedy 
Space Center; a series of workshops 
for discussing human factors issues 
in aviation and space vehicle 
maintenance is being conducted. 
The first workshop was held at Ames 
Research Center in September 1966; 
it focused on error analysis (inci- 
dents, accidents, mishaps, and close- 
calls). This workshop facilitated the 



free exchange of human factors 
information [between aircraft and 
shuttle maintenance operations. 

Point of Contact: B. Kanki 
(650) 604-5785 
bkanki@mail.arc.nasa.gov 



Aircraft Separation 
Risk Model 

Mary M. Connors 

Increased demand for air travel 
translates into a need to accommo- 
date more aircraft in the terminal 
airspace. Separations between 
aircraft pairs must be small enough 
to be efficient while remaining 
sufficiently large to be safe. Present 
separation standards are considered 
safe for the equipment and condi- 
tions that presently prevail. It is 
possible that new technologies could 
result in maintaining the present 
safety level while permitting the 
distances between aircraft pairs to 
be reduced. What is needed is a 
quantitative method of evaluating 
safety. The objective of this research 
is to develop a computer model that 
will provide this link between 
aircraft separation and a quantitative 
method of assessing safety risk. 

The model, called the Reduced 
Aircraft Separation Risk Assessment 
Model (RASRAM), evaluates safety 
risks for a variety of flight scenarios 
relating to final approach, landing, 
and rollout for parallel and single 
runways. The basic approach of 
RASRAM is to quantify the risk 
associated with current separation 
standards and to then compare it 
with that for reduced separation 



(J I n H A I, (] [ \ 1 L A V 1 A T [ON / Safety 



MD = miss distance 



Blunder path 



Blunder aircraft 



3 Glide path 




Fi^. L Parallel approach lateral sepan-ition geometry. 



operations during instrument 
meteorological conditions, consider- 
ing procedural and technological 
changes. The research is being 
performed tor Ames Research Center 
as an integral part of NASA's Termi- 
nal Area Productivity (TAP) program, 
and in coordination with the Federal 
Aviation Administration. 

RASRAM includes the following 
two scenarios that measure the effect 
of separation on safety: lateral 
separation for parallel approaches; 
and in-trail separation during single 
runway operations, accounting for 
runway occupancy and wake vortex 
effects. The figure illustrates the 
parallel approach scenario, for 
which independent approaches to 
parallel runways constitute the 
operational context. The primary 
separation criterion is the distance 
between the runway centerlines for 
the two approach paths. The defin- 
ing characteristic of the scenario in 
the model is a blundering aircraft 
that strays from its own final 
approach, crossing the path of the 
other approach stream. The safety of 
the scenario (for current operations) 



is determined (1 ) by the performance 
of the controller in detecting the 
blunder and issuing breakout 
instructions to the evader aircraft, 
and (2) by the performance of the 
pilot and aircraft in completing the 
evasive maneuver. RASRAM models 
all of these effects and quantifies the 
safety of the operation. The standard 
risk measure is the probability that 
the blundering aircraft will approach 
within 500 feet of an aircraft in the 
other approach stream. 

NASA's TAP Program anticipates 
procedural changes for terminal-area 
operations along with the introduc- 
tion of new technologies. These new 
technologies include Differential 
GPS (DGPS), Automatic Dependent 
Surveillance (ADS-B), and several 
technologies under development by 
the NASA TAP program: the Center- 
TRACON Automation System 
(CTAS), the Aircraft Vortex Spacing 
System (AVOSS), the Dynamic 
Runway Occupancy Measurement 
(DROM), and the Airborne informa- 
tion for Lateral Spacing (AILS). 
RASRAM begins the process of 



quantifying the safety risk associated 
with the effects of these technologies 
on separation standards. The model 
can also be used to provide a 
relative comparison of the safety of 
proposed new procedures with the 
safety of current operations and 
technologies. Using RASRAM, the 
safety of further separation reduc- 
tions for parallel approaches that 
rely on new technologies can be 
analyzed and compared with current 
procedures. Similarly, the safety of 
reductions of in-trail separation can 
be compared with current proce- 
dures. Although it is being devel- 
oped for initial application to final 
approach and landing, the basic 
approach to modeling separation 
risk has direct application to all 
phases of flight. The potential 
application of these models to en 
route and approach airspace is being 
undertaken in the Advanced Air 
Transportation Technology program 
at Ames. 

Point of Contact: M. Connors 

(650)604-6114 

mconnors@mail.arc.nasa.gov 



10 



G I H A [, C I \' 1 L A V ! A T ION / Siifely 



A E RO SAVT I C S AS D S P AC E T R A S S P R T A T I S 
T E C H S L G Y E \ T E R P R IS E 



The Final Approach 
Spacing Tool 

Tom Davis 

The Center/TRACON Automa- 
tion System (CTAS) is an air-traffic 
control automation system under 
development at Ames Research 
Center. The system Is designed to 
support increasing demands for 
capacity and efficiency in the 
National Airspace System. The 
terminal-area component of the 
system, the Final Approach Spacing 
Tool (FAST), is designed to assist 
terminal air-traffic controllers In 
efficiently managing and controlling 
arrival air traffic for the last 40 miles 
of flight down to the runv^ay. The 
FAST system issues the controllers 
a series of sequencing, runway, 
heading, and speed advisories to 
achieve an efficient flow of traffic 
that increases airport capacity, 
reduces delays, and reduces control- 
ler workloads. 

The objective of the research 
and development effort is to test the 
FAST system operationally in a series 
of phased functionality enhance- 
ments at the Dallas/Fort Worth 
TRACON (Terminal Radar Approach 
Control) (see figure (see Color 
Plate 1 in the Appendix)). The 
operational field testing of the FAST 
system allows researchers to further 
develop and assess the system with 
the end-users, that is, with the air- 
traffic controllers. The assessments 
include a series of observations with 
the FAST system operating in a 
shadow mode on live traffic data, 
real-time simulation evaluations of 



the FAST system at Ames and at the 
FAA Technical Center, and a limited 
operational assessment of the 
FAST system functionalities at 
Dallas/Fort Worth. 

Development and field testing of 
the FAST system achieved a signifi- 
cant milestone during FY96; the 
''Passive" FAST functionality, which 
includes the sequence and runway 
advisories, underwent operational 
testing at the Dallas/Fort Worth 
TRACON. The tests, which were 
conducted in cooperation with the 
FAA, the National Air Traffic Con- 
trollers Association, and the Airline 
Transport Association, began in 
January 1996 and continued through 
July 1 996. The objectives of the test 
were: (1) to confirm the technical 
performance of the Passive FAST 
functionalities, (2) to assess the 
airport delay and capacity benefits 
of Passive FAST, and (3) to assess 
the controller workload benefits of 
Passive FAST. The testing demon- 
strated an arrival rate improvement 
at Dallas/Fort Worth of 1 3%, a 
departure queue backlog reduction 
of 9%, and an overall increase In 
total airport operations (arrivals and 
departures) of 13%, all with no 
increase In taxi times and with 
little or no increase in controller 
workload. The system has received 
overwhelmingly positive support 
from all participants. 

The Active FAST system 
functionalities, including speed and 
heading advisories, remain to be 
developed and operationally evalu- 
ated. The Active FAST system will 
undergo operational testing similar 



to that of the Passive FAST before a 
national deployment system is 
specified. 

Point of Contact: T. Davis 
(650) 604-5438 
tdavis@mail.arc.nasa.gov 



Air/Ground Integration 

R. Slattery 

The Center/TRACON Automa- 
tion System (CTAS) was developed 
for use in the air traffic-control 
environment. As part of the Terminal 
Area Productivity Air Traffic Man- 
agement program, research has 
begun to expand CTAS to coordinate 
with aircraft that are equipped with 
Flight Management Systems (FMSs) 
using data-link. A large portion of 
the commercial aircraft fleet is so 
equipped and many older aircraft 
are being retrofitted with an FMS. 
However, busy air-traffic controllers 
do not often allow pilots to fully 
utilize the FMS. An FMS calculates 
the most efficient trajectory, thus 
saving fuel, and then follows the 
trajectory very accurately, increasing 
trajectory prediction accuracy. Thus, 
If the CTAS helps the controllers take 
advantage of the FMS trajectory, the 
performance of the airspace system 
should be improved. 

The objectives of the research 
and development effort are to 
quantify the actual benefits of the 
air/ground system with both control- 
lers and pilots in the loop. Theoreti- 
cal benefits have been studied, 
under the assumption that the 
trajectories are followed with the 



Ci I- H A !. Civil. A V I A T I N / Safety 



11 



maximum accuracy possible for the 
FMS. These benefits may be dimin- 
ished, however, by the extra 
workload imposed on the controllers 
and pilots; also, they may not be 
realizable because of operator error. 

A piloted simulation was 
performed using the CVSRF 747-400 
simulator. Since the simulator was 
not equipped with data-link at the 
time, the routes were defined as 
stored arrivals in the FMS (an 
example is shown in the figure). 
The pilots were issued FMS route 
changes, while still performing the 
rest of their normal duties. For a 
baseline comparison, equivalent 
routes were flown using current 
controller clearances. The aircraft 
was flown in the Dallas/Fort Worth 
terminal area, starting at the two 



KINGG 

REFLE2:Crossat210K 

IAS. Cross at 4,000* 



eastern arrival gates into the 
TRACON (Terminal Radar Approach 
Control) and finishing at the runway. 
The accuracy of the horizontal route 
increased for the FMS cases, but the 
vertical predictability was lower. 
This was because the vertical portion 
of the FMS route is entered as 
crossing altitudes at way points. As 
long as the pilots meet the required 
altitude, the point at which they 
begin their descent and the way in 
which they make the descent are 
open. Offsetting the achievable 
accuracy, the pilots felt that the FMS 
route changes produced far more 
workload with current generation 
FMS systems than did current 
procedures. They also expressed 
concerns about the amount of 
heads-down time spent loading the 



O- 



17LGT 



JIFFY 
Cross at 170K IAS 



REFLE2 NOVEL 

i 



t 

O 



REFLE1 



O SWIFT 

REFLE1:Crossat210K 
IAS. Cross at 4,000' 



REFIL 

Cross at 21 OK IAS. 

Cross at 11,000' 



6 



SEAGO 

Cross at 250K IAS. 

Cross at 11,000* 



SCY 




Fig, L Scurry FMS arrivals, RFFLFl and RFFLF2. 



trajectory while in the busy terminal 
area. There were also more errors in 
loading and in following the FMS 
route, though these should be 
reduced by data-link and by inter- 
face changes to the FMS. 

Point of Contact: R. Slattery 
(650) 604-5435 
rslattery@mail.arc.nasa.gov 



The Traffic Management 
Advisor 

Harry N. Swenson 

The growth of commercial air 
travel within the United States has 
put a severe strain on the nation's air 
traffic capacity. This, coupled with 
the ''hub-and-spoke" procedures 
used by the major air carriers and 
the marketing requirements for 
aircraft to take off and land at 
optimum times, has required an 
improvement in the Air Traffic 
Control System. The Center- 
TRACON Automation System (CTAS) 
is a decision-support concept being 
developed to improve airport 
capacity and to reduce delays while 
maintaining controller workload at 
a reasonable level. The extended 
terminal-area component of the 
system is the Traffic Management 
Advisor (TMA). The TMA is a time- 
based strategic planning tool that 
provides traffic management coordi- 
nators and en route air traffic 
controllers the ability to efficiently 
optimize the capacity of a demand- 
impacted airport. The TMA consists 
of trajectory prediction, constraint- 
based runway scheduling, traffic 



12 



Glob 



A I C I \' 1 i. A V 1 A T I N / Safely 



Aero \ a i r i c s a s d Space T r a s s p o r t a t i o > 

T E C H S I G Y ESTERPRISE 



flow visualization, and controller 
arrival sequence, time, and delay 
advisories. 

The TMA and all other CTAS 
tools are being developed to demon- 
strate user benefits both to air-traffic 
controllers and commercial air 
carriers. Air-traffic control is a very 
complex multivariable problem 
requiring teams of highly skilled 
individuals to safely and efficiently 
move traffic from terminal departure 
point to terminal arrival. This 
extended complexity requires that 
automation aids designed to assist 
these teams be proven and validated 
in actual operations. The objective 
of the TMA 1 996 research and 
development effort v^as focused 
toward an operational evaluation 
at the Ft. Worth Air Route Traffic 
Control Center, Ft. Worth, Texas. 
The evaluations also focus research 
into areas and concepts that provide 
user benefits. 

The TMA achieved several 
major milestones in this focused 
effort toward operational evalua- 
tions. The first was controller and 
operational hardware in-the-loop 
simulations at the William J. Hughes 
FAA Technical Center with the 
National Air Traffic Controllers 
Association (NATCA) Ft. Worth 
Center System Design Team. This 
was followed by the installation and 
checkout of the TMA hardware/ 
software components at the 
Ft. Worth Center, the nation's fifth 
busiest Center, which manages and 
controls traffic into the second 
busiest airport in the world (Dallas/ 
Ft. Worth). Upon completion of the 
system installations, on-site shadow 
evaluations and nighttime simula- 



tions were conducted. Subsequently, 
the formal operational evaluations 
were conducted. The TMA 
demonstrated delay reductions of 
1-2 minutes per aircraft, as well as 
a significant reduction in controller 
workload. The original plan was to 
remove the TMA upon completion 
of the formal evaluation, but, based 
on the benefits achieved, the FAA, 
the Air Transport Association, and 
NATCA requested that NASA 
operationally support the TMA on a 
continuous basis. NASA has com- 
plied with that request, and the TMA 
has been used operationally since 
the completion of the formal evalua- 
tions. Since the evaluations, a 1 0% 
capacity increase has been attributed 
to TMA operations. 

Point of Contact: H. Swenson 
(650) 604-5469 
hswenson@mail.arc.nasa.gov 



Surface Movement 
Advisor 

Brian J. Glass 

Recurrent delays during depar- 
ture taxiing at large airports have 
become commonplace with the 
prevalence of "hub-and-spoke" 
airline operations, as large numbers 
of aircraft attempt to land, taxi, be 
serviced, taxi, and depart, all within 
60-90 minute "banks." Airfield 
tower controllers strive to avoid 
imbalances and bottlenecks by 
integrating data from visual cues 
and from a variety of other sources. 
However, lengthy, imbalanced taxi 



queues are evidence that although 
sufficient and safe, the controller's 
mental planning process {given 
current data sources) is not necessar- 
ily optimal. By providing data fusion 
and automated optimal taxi plan 
advisories to the ground controller, 
the controller can operate with 
improved data-gathering and 
planning capabilities. Improved 
dynamic taxi routing, and hence 
smoother airport operations with 
less surface taxi delay should result. 
Eventual national implementation 
of the Surface Movement Advisor 
systems (SMA) at the 1 3 largest U.S. 
airports is projected to save users at 
least 5% of the $1 .6 billion annual 
ground-delay costs incurred as a 
result of inefficient taxi and runway 
queuing — based on recent FAA 
simulations and Air Transport 
Association cost figures. 

The SMA is a series of succes- 
sive airfield data systems that is 
being developed as a joint effort 
between Ames Research Center and 
the Federal Aviation Administration 
(FAA). The first proof-of-concept 
SMA (Build-1) electronically con- 
nects the air-traffic control, airline, 
and airport operations users at 
Atlanta-Hartsfield airport (ATL) to 
facilitate information-sharing and 
data fusion. SMA Build-1 has been 
in use on a daily basis at ATL by 
airlines and airport ramp towers 
since June 1996 and was brought 
online in the FAA control tower 
in September 1 996. The figure 
shows an overview of the SMA 
system that was deployed to Atlanta 
in 1996. 



G L B A L G I V 1 I A V 1 A T ! N / Safe!}' 



13 



VatuJet 
Lockheed Other airlines 



Delta 
Airlines 



Real-time 
radar data 



City of Atlanta 
airport authority 




Flight plans 



Performance Prediction Airport operations Statistical 

histograms algorithms procedures analysis 



Weather 
data 



Real-time aircraft 
status updates 



Airline 
schedules 



Fig. 1. Overview of the SMA system deployed to Atlanta in 1996. 



Running in a commercial 
transaction-processing database on 
an off-the-shelf UNIX server, the 
SMA software currently contains its 
own tracking, time estimation, data 
fusion, monitoring and prediction 
software modules. SMA furnishes 
data as an ASCII text stream to 
airline servers, or as separate 
X/Motif-based touchscreen-capable 
graphical-user interfaces for the FAA, 
airline, and airport operations 
personnel. 

Point of Contact: B. Glass 

(650)604-3512 

bglass@mail.arc.nasa.gov 



Simplified Vision Models 
for Display Quality 
Assessment 

Al Ahumada 

Detection and recognition of 
obstacles (aircraft, trucks, etc.) is 
crucial for safe aviation operation in 
the terminal area. Vision models can 
make an important contribution to 
display design by allowing computa- 
tional predictions of the visual 
adequacy of potential designs for 
obstacle detection and recognition. 
The first figure shows a pair of 
simulated display images, one 
(panel b) with an obstacle on the 
runway and one without. Investiga- 
tors have been measuring the ability 



of human observers to make such 
discriminations and have been 
constructing and testing computa- 
tional models to predict their 
abilities. 

Image discrimination models 
that have been used as image-quality 
metrics range in complexity from 
single filter models to multiple- 
channel models with channels that 
are selective in spatial frequency and 
orientation. The simple filter models 
can be thought of as representing the 
visual information at precortical 
levels of the visual system. These 
models can predict the variations in 
the visibility of targets that occur as 
the target spatial frequency changes. 
The multiple-channel models 



14 



C; !. {) H A L C 1 V ! L A V i A T ION / Safety 



a f r s a i t i € s a s d s p a c e t r a s s p r t a t i s 
Techsology enterprise 




Fig. L Panel (a) is a 128 x 128 pixel 
gray-scale digital image of a 
simulated airport runway scene. 
Panel (b) shows the runway scene 
with an aircraft obstacle. 



simulate the orientation and spatial 
frequency selectivity of cortical cells. 
Because the channel outputs are 
nonlinear, the channel models 
predict masking of targets in high- 
contrast image regions. For example, 
the channel models outperform 
contrast-sensitivity filter models in 
predicting the detectability of targets 
in a natural background. 

Some recent vision models 
include between-channe! interac- 
tions that allow the models to 
predict masking from image compo- 
nents exciting different channels 
from those responding to the target. 
These models have even greater 
computational complexity. Assum- 
ing homogeneity of these interac- 
tions leads to a simple contrast 
masking correction. The second 
figure diagrams the image discrimi- 
nation model that results from 



Bacl<ground 




RMS 



J'4'' 



Masking 



lui^)' 



Euclidean 
distance 







r 



Fig. 2. Schematic of the simple image discrimination model. At the left, the 
upper image is the background image and the lower image is the target-plus- 
background image. After the luminance images are converted to contrast 
images (a step not illustrated), the contrast images are filtered by a contrast 
sensitivity function (CSF). On the top path, the masking parameter c is the 
rms value of the background-filtered contrast values t]. On the bottom path, 
the corresponding filtered values from the image are differenced and the 
vector length of these dj values is divided by the contrast factor to get the 
predicted number of just-noticeable-differences between the two images. 



applying this correction to a 
contrast-sensitivity filter model. In 
comparing the predictions of this 
model with those of the complex 
models, it is found that the simple 
model predicts human visual 
detection performance just as well. 

Point of Contact: A. Ahumada 
(650) 604-6257 
aahumada@mail.arc.nasa.gov 



Perceptually Tuned Visual 
Simulation 

Mary K. Kaiser 

Human factors engineering is 
required to improve the quality of 
visual displays in aerospace systems. 
Advanced computer-generated 
imagery (CGI) systems are used to 
create compelling visual displays for 
navigation/control systems, vehicle/ 
system simulation, telerobotics, and 
scientific visualization applications. 
The quality of these displays can 
affect the safety and productivity of 
flight and ground-based operations. 
Inevitably, the realism of these 
displays is constrained by limitations 
in CGI hardware and software, 
especially if images need to be 
generated in real-time. Despite 
rapid advances in image-generation 
technology, human operators desire 
more realistic, higher-fidelity 
displays; it is likely that this demand 
for improved fidelity will continue 
for the foreseeable future. 

Research is being conducted 
to examine techniques aimed at 
reducing the computational cost 
required to achieve a desired level of 
image quality and frame rate. These 



(i I li A [. (] I V 1 L A \ 1 A T ION / Safety 



15 



techniques exploit principles of 
visual processing to reduce the 
computational load. This nnulti- 
disciplinary research involves a 
collaboration among research 
scientists at the Ames Research 
Center; professors in computer 
science at Carnegie Mellon Univer- 
sity and in psychologyA^iomedical 
engineering at the University of 
Virginia; and designers and engi- 
neers at various industry sites. 



A set of techniques has been 
developed for rendering images with 
enhanced apparent resolution. 
NASA has applied for a patent for 
this process (NASA Case: ARC 
1 2080-1 ). In using this process to 
create stereo displays, images with 
different resolutions are shown to the 
two eyes; examples of this technique 
are shown in the figures below. The 
first figure demonstrates varying 
texture complexity; the second 




Fig, L Only the image presented to the left eye in this stereo pair has texture 
mapping. Nonetheless, the resulting stereo percept appears textured. 




Fig. 2. Another example of ''hi-lo'' stereo images. Here the left image is 
created with far more detail (that is, a larger number of polygons is used to 
model the object) than the one on the right. The resulting stereo perception 
appears high resolution. 



demonstrates varying polygonal 
complexity. The resulting fused 
image appears to possess the higher- 
resolution detail. User studies 
conducted with these ''hi-lo'' stereo 
displays indicate that people are 
able to extract stereo-specified depth 
with these displays about as well as 
they do with traditional (''hi-hi") 
displays. Further, no interference 
with normal stereo vision results 
from exposure to the hi-lo displays. 
Present efforts focus on further 
evaluation of the algorithms, the 
extension of techniques to larger 
object classes, and the development 
of general izable tools suitable for 
inclusion in a graphics modeling 
toolbox. Coordination with hard- 
ware and software developers seeks 
to maximize the utility of these 
techniques and to ensure compat- 
ibility with hardware architecture. 

The development of a number of 
rendering techniques that signifi- 
cantly enhance the performance of 
graphics systems is expected. The 
goal is to both extend the upper 
range of graphical rendering perfor- 
mance and to enable lower-end 
systems to produce visual imagery 
that currently can be produced only 
with high-end systems. 

Point of Contact: M. Kaiser 
(650) 604-4448 
mkaiser@mail.arc.nasa.gov 



16 



Global Civil Aviation / Safeiy 



AEROSAi'TfCS A S D S P A C E TrASSFORTATIOS 
T E C H S L G Y E S T E R P R I S E 



Entropy Masking in Visual 
Displays 

Andrew B. Watson 

As part of NASA's Human 
Exploration and Development of 
Space Enterprise and Aeronautics 
and Space Transportation Technol- 
ogy Enterprise, the Vision Science 
and Technology Group at Ames 
Research Center is doing research on 
the human factors of visual displays 
and visual communication systems. 
A primary thrust of this effort is to 
develop computational models of 
human visual performance that can 
be used in the engineering design of 
imaging and display devices. A first 
goal of human vision models is to 
predict visibility of arbitrary targets, 
especially against noisy or cluttered 
backgrounds that are typical of real 
imaging environments. This problem 
is central to applications such as 
image compression, flat-panel 
display design, sensor fusion, and 
design of helmet-mounted and other 
head-up displays. To meet this 
challenge, a program has been 
initiated involving experiments on 
visual masking: the interference of 
the background with the detection of 
visual targets. 

As a result of this research, a 
new category of visual masking has 
been identified; it is called entropy 
masking. Previous research over 
many decades had identified two 
principal varieties of visual masking: 
contrast masking and noise masking. 
The former refers to a reduction in 
visual sensitivity resulting from a 
passive adaptive process. In effect, 
the gain of visual neurons is turned 
down when there is a large amount 
of pattern stimulation, in order to 



prevent saturation of the limited 
dynamic range of the neural 
response. Noise masking refers to 
an inevitable reduction in visibility 
when a random component is added 
to the visual stimulus. Entropy 
masking depends on neither gain 
control nor randomness. Instead, it is 
due to the observers lack of knowl- 
edge about the obscuring back- 
ground, and it may be ameliorated 
by increasing that knowledge 
through learning. 

The following experiments 
illustrate the essentials of entropy 
masking. The observer attempted to 
detect the Gabor-function target 
pictured in part (a) of the first figure; 
the target was obscured by the 
addition of one of the masks shown 
in parts (b) and (c). The mask shown 
in part (b) is a cosine luminance 
grating with the same spatial fre- 
quency as the target, and that shown 
in part (c) is a sample of one-octave, 
bandpass noise with a center spatial 
frequency equal to that of the 
Gabor-function target. The two 
masks were equated in contrast 
energy so that they would have 
equal effects on the contrast gain- 
control system, that is, equivalent 
contrast masking. 

On each trial, the observer 
received two brief presentations on 
a cathode-ray-tube monitor. One 
contained a background only, the 
other contained a background plus 
the target. The observer tried to 
identify the presentation containing 
the target. The contrast of the target 
was varied from trial to trial in order 
to estimate the contrast required 
for 827() correct performance. To 
manipulate the degree of random- 
ness, the bandpass mask was 
(1 ) new on each presentation 




Fig. 1. Visual target (a) and two 
masks: cosine grating (b) and 
bandpass noise (c). Each mask is 
shown with the target added. 



G I B A I C i \ I L A \ ! A T [ON / Safety 



17 




Fig. 2. Target thresholds as a function of mask condition. 



(random condition), (2) the same 
for both presentations of a trial but 
new for each trial (twin condition), 
or (3) always the same (fixed 
condition). 

The results are shown in the 
second figure. The cosine mask 
elevates the target threshold by a 
small amount (about 3 decibels), but 
the random condition produces 
much more masking. However, the 
twin condition shows that this 
masking is not due to randomness, 
since the masks are identical in the 
two presentations of each trial. The 
fixed condition shows that with 
learning (640, 21 76, and 3584 trials 
were collected from the three 
observers), a significant reduction in 
masking is obtained. Indeed, at the 
end of learning, the threshold is 
similar to that for the cosine condi- 
tion. It can be concluded that the 



masking effect of the bandpass noise 
is due to the observer's ignorance 
about its structure (its entropy). 

It is thought that entropy mask- 
ing Is the dominant form of masking 
in many applied situations. Conse- 
quently, it is important to verify its 
existence and properties in order to 
enable optimal design of visual 
communications systems. This work 
moves NASA closer to its goal of 
developing a set of general human 
factors engineering tools that will 
enhance the safety and effectiveness 
of its Space and Aeronautics 
missions. 

Point of Contact: A. Watson 

(650)604-5419 

abwatson@mail.arc.nasa.gov 



Vertical Motion Simulator 
Advanced Simulator 
Network 

William B. Cleveland 

The Advanced Simulator 
Network (ASN) is an integration of 
newer and higher-performance host 
computers and real-time data 
acquisition networks into the 
Vertical Motion Simulator (VMS) 
complex at Ames Research Center. 
The system provides significantly 
better performance while retaining 
the powerful functionality of the 
previous system. As shown in the 
figure, the system consists of the data 
network, instrumentation input/ 
output (I/O), and a host computer 
together with its interface to the 
network. In a simulation setting, the 
pilot's controls are input through the 
Cockpit CAMAC (Computer Auto- 
mated Measurement and Control) 
I/O, passed via the Data Highway 
and VME interface to the host 
computer where the aircraft's 
response is computed and returned 
via the network highway elements to 
the motion, cockpit display, and 
laboratory display devices. 

The performance increase of the 
operational systems is dramatic. By 
themselves the computers are twice 
as fast as the computers of just 
4-5 years ago. Testing of the new 
CAMAC instrumentation shows a 
factor of 10 speed-up in perfor- 
mance over that of the old hardware. 
These improvements reduce time 
delays that cause a simulated aircraft 
to lose stability In comparison with 
the real aircraft. When there is no 
real aircraft for comparison, incor- 
rect conclusions are often made 



18 



(J I. {) B A I. Civil Aviation / Safely 



A E RO S A IT I C S A S D S P A C E T R A \ S P R T A T I S 
T E C H S L 6 Y E \ T E R P R I S E 




Data "highway" 



k 



^ 



Laboratory 



Fig. 1. Advanced Simulator Network. 



when the simulator uses poor 
equipment or techniques. Although 
the VMS system can operate at 
1 000 hertz when only I/O is per- 
formed with the motion, laboratory, 
and cockpit subsystems, adding the 
typical aircraft model allows a 
200-hertz operation. Most simula- 
tions are near the 60-hertz rate of the 
associated graphics generators, but 
the ASN system performance reserve 
will allow larger model options and 
I/O rates when needed. 

Early in FY96, the first of the 
systems was installed into an 
Interchangeable Cab fixed-base 
simulation laboratory, where it was 
tested and perfected before it was 
installed into the VMS. Before the 
end of FY96, the VMS system had 
been used in five simulations 
including astronauts flying the space 
shuttle vehicle. 

Point of Contact: W. Cleveland 
(650) 604-5352 
wcleveland@mail.arc.nasa.gov 



Motion 
CAMAC I/O 



Cockpit 



y 



Intelligent Aircraft Control 
System 

Charles C. Jorgensen 

The intelligent aircraft control 
system provides a safety-oriented 
flight control technology that can 
efficiently deal with off-nominal and 
unforeseen changes in aircraft 
systems or operating environments. 
The effort accomplishes reconfig- 
urable flight control under damage 
scenarios by using a modified F-1 5 
aircraft's (ACTIVE) differential canard 
and vectored thrust capabilities. This 
effort is a four-year cooperative 
research contract begun in 1995. 
It comprises three phases: Phase I, 
create neural nets (NNs) that model 
stability and control derivatives in 
the pitch, roll, yaw, lift, and side 
force axes of the F-1 5 ACTIVE 
aircraft in 1995-96 (see the figure 
(see Color Plate 2 in the Appendix)); 
Phase II, develop on-line NNs to 
process aircraft sensor data in real- 
time and update the stability and 
control derivatives in 1997; and 



Phase ill, flight demonstration of 
real-time NN control and accident 
reconfiguration (1997, 1998). 

In 1996 major progress was 
made. Phase I NNs were generated 
that match the ACTIVE aerodynam- 
ics and were programmed into the 
F-15E computer. Three Phase I NNs 
were produced: (1) the Leavenburg- 
Marquardt network at Ames 
Research Center (ARC), (2) an active 
selection net at McDonnell Douglas 
Aircraft Corporation (MDAC), and 
(3) a neural fuzzy (ANFIS) at 
Tennessee State University. 

Four second-generation on-line 
networks were developed for 
Phase II: Radial Basis Networks 
(MDAC), Recursive Higher Order 
Neural Network (Washington 
University), Feedback Linearization 
Neural Net (ARC), and Dynamic Cell 
Structures (ARC). 

Two on-line aerodynamic 
derivative estimation methods were 
developed by MDAC: Kalman 
Filtering/Recursive Prediction and 
Locally Weighted Regression. Three 
Linear Quadratic Regulator Methods 
were designed by MDAC: (1) Robust 
Servo, (2) Setpoint Regulator, and 
(3) SOFFT, Langley Research Center. 
A full-scale parameter sensitivity 
analysis was begun for performance 
envelope, stability, and robustness 
using the new techniques. A real- 
time Ricatti equation solver was 
developed (MDAC), and a new 
Virtual Reality test environment was 
developed by ARC using a nonlinear 
F-1 5E model. 

Point of Contact: C. Jorgensen 
(650) 604-6725 
cjorgensen@mail.arc.nasa.gov 



^i 1. n A I, C I V I i. A \' I A r I o n / Safety 



19 




Facilitating User Route 
Preferences in En Route 
Airspace 

Bob Vivona, Mark Ballin, 
Steve Green, Ralph Bach, 
Dave McNally 

Under the current air-traffic 
systenn, aircraft generally must fly on 
fixed routes. On the other hand, the 
Air Transport Association estimates 
that its member airMnes could save 
$1 .2 billion per year if they were 
able to fly direct routes. However, 
relaxing existing routing restrictions 
will require new conflict-prediction 
and resolution automation to help 
controllers maintain required 
separation while minimizing their 
effect on their workload. Although 
some automation now exists (for 
example, Conflict Alert, which uses 
a 2-minute prediction based on 
previous radar hits), for the most part 
air-traffic controllers detect conflicts 
by visually monitoring radar tracks 
and, when necessary, issuing 
resolution advisories to aircraft to 
avoid loss of separation (5 nautical 
miles horizontal, 2000 feet vertical). 
Conflict-prediction and resolution 
tools for controllers are being 
developed under the Advanced Air 
Transportation Technology program 
at Ames Research Center. The 
objective of this work is to develop 
an operational concept and associ- 
ated software for use of conflict- 
prediction information in order to 
generate minimum-cost resolution 
advisories to facilitate increased 
user route preference in en route 
airspace. 



A wide range of traffic environ- 
ments is found in the National 
Airspace System. At one extreme is 
en route airspace away from busy 
terminal areas consisting of aircraft 
generally not constrained by traffic 
or weather. If traffic density is low, 
very little ground-based air-traffic 
control intervention is required. At 



the other extreme is the en route 
area near busy terminal areas where 
aircraft must be merged for spacing 
and sequencing. The system is 
supplied with real-time tracking 
radar and flight plan information 
from the air-traffic control facility's 
host computer. The Airspace Tool 
(AT) is designed to achieve user 



Radar tracks, Flight plans 



' f 


Conflict resolution 
advisories 


i 


Airspace 
tool 


Sector 
tool 








Clearance 



information 



Aircraft 



Fig, h Operational concept for user-preferred routing in en route airspace. 




Fig. 2. Sample airspace with four sectors showing two typical conflict 



scenarios. 



20 



(1 I \\ \ 1. (^ I \ I 1. 



A V 1 A T ! N / AffordahiUty 



A E R ^ A I T ! C S A S D S F A C E TrASSPORTATIOS 
T E C H \ L C Y E S T E R P R I S E 



efficiency for long tinne-horizon 
(15-25 minutes), generally inter- 
sector conflict-predictions by 
generating resolution advisories 
based on minimum cost deviation 
from user-preferred routes. Reduced 
controller workload is also achieved 
through inter-sector conflict resolu- 
tion. The Sector Tool (ST) facilitates 
controller planning within a sector 
for short time-horizon conflict- 
prediction, trial plan (''what if) 
solutions, and transition to the 
terminal area (see first figure). 

The second figure shows a 
simple example of en route airspace 
with four sectors and two sample 
conflict scenarios. In both cases a 
minimum-cost resolution advisory 
is computed by the AT. For the case 
in which the conflict is predicted 
across sector boundaries, the AT 
passes a suggested advisory to the 
appropriate sector. Each resolution 
is checked by the sector controller 
with the aid of the ST, and, if 
acceptable, is issued to the aircraft 
by the controller. 

The Center-TRACON Automa- 
tion System software is being used 
as a baseline for development, 
simulation, and field testing of the 
operational concept. 

Point of Contact: D. McNally 
(650) 604-5440 
dmcnaily@mail.arc.nasa.gov 



Conflict Probability 
Estimation for Free Flight 

Russell A. Paielli, Heinz Erzberger 

The Air Transport Association 
estimates that the U.S. airline 
industry could save several billion 



dollars annually in direct operating 
costs if the Federal Aviation Admin- 
istration (FAA) would relax routing 
restrictions and adopt a policy of 
''free flight." The safety and effi- 
ciency of free flight will benefit from 
automated conflict-prediction and 
resolution advisories to help air- 
traffic controllers maintain the 
required separation between aircraft 
while minimizing the effect on 
trajectories. Trajectory prediction 
is less certain the further in advance 
a prediction is made, however, 
so conflicts cannot be predicted 
exactly. A method is needed, 



therefore, to estimate the probability 
that a conflict will occur between a 
given pair of aircraft. This conflict- 
probability estimate is important for 
determining both how and when to 
optimally resolve conflicts. 

An error model and an algo- 
rithm have been developed to 
determine conflict probability. Based 
on analysis of actual air-traffic data, 
the trajectory-prediction errors are 
modeled as being normally distrib- 
uted (Gaussian), with the along-track 
error growing with prediction time 
as a result of wind modeling error, 
as shown in the first figure. The two 



Prediction time 



Path 

crossing 

angle 




Predicted 
trajectories 



Minimum 
predicted 
separation 



(I 



« 



Prediction 

error 

ellipse 



Predicted time 



Fig. 1. Trajectory prediction error ellipses. 



G L H A I, C I V ] L Aviation / Affordabtlity 



21 



Combined 
error ellipse 




Relative velocity 



Extended conflict zone 



Fig. 2. Encounter geometry. 



error covariances for an aircraft pair 
are combined into a single equiva- 
lent covariance of the relative 
position. The second figure shows 
the combined error ellipse (based on 
the combined covariance) centered 
on one aircraft and the circular 
conflict zone centered on the other. 
The extended conflict zone is the 
projection of the circular conflict 
zone in the direction of the relative 
velocity. In two dimensions the 
conflict probability is the volume 
under the probability-density 
function that is within that extended 



conflict zone. In three dimensions 
the basic concept is similar but more 
difficult to illustrate. 

A coordinate transformation 
based on the Cholesky decomposi- 
tion C'square-root" factorization) of 
the combined covariance is used to 
derive an analytical solution for the 
conflict probability. If the Gaussian 
error model were exact, this solution 
would be exact for level flight and a 
good approximation for nonlevel 
flight. This analytical solution is 
computationally far more efficient 
than an alternative solution based on 



numerical integration. Efficiency is 
important because the algorithm is 
intended to run in real time. The 
high efficiency also makes the 
algorithm a convenient toot for 
analyzing the effects of various 
parameters. 

The algorithm was programmed 
in C-H-i- and successfully tested by 
Monte Carlo simulation over a wide 
range of encounter geometries. It has 
also been integrated into the Center/ 
TRACON Automation System 
(CTAS). CTAS is a large software 
system developed at Ames Research 
Center to help air-traffic controllers 
do their job more efficiently and 
with less stress. In this context, CTAS 
predicts the trajectories, determines 
the states of the aircraft at the 
predicted point of minimum separa- 
tion, and sends that state information 
to the conflict-probability algorithm. 
The probability results are then used 
by CTAS to determine if a predicted 
conflict should be displayed to the 
controller. The system is currently 
being tested on both live and 
recorded air-traffic data that are sent 
to Ames on a dedicated line from 
Denver Center. 

Point of Contact: R, Paielli 
(650) 604-5454 
rpaielli@etna.arc.nasa.gov 



22 



G 1, h A L C ! V ! I, A V 1 A T i N / Ajjo v dii i) Hi ty 



Aero \ \ i i i f s ,\ s d Space Trassportatio s 

T F C H S t G y E S T E R P R I S E 



Conflict Prediction 
Algorithms: Initial 
Field Test 

Dave McNally, Bob Vivona, 
Karl Bilimoria, Gerd Kanning, 
Steve Green, Ralph Bach, 
Allan McCrary, Ed Lewis 

The Federal Aviation Adminis- 
tration (FAA) is responding to the 
needs of the aviation connmunity by 
exploring the concept of ''free flight'' 
for aircraft flying under instrument 
flight rules. In free flight, en route 
aircraft fly user-defined trajectories 
(defined by pilots or airlines) v^ith 
only minimal air-traffic control (ATC) 
adjustments to avoid restricted 
airspace and loss of separation from 
other aircraft. To facilitate user- 
preferred routing in all en route 
environments, an extension of the 
ground-based Center-TRACON 
Automation System (CTAS) is being 
developed under the Advanced Air 
Transportation Technology program 
at Ames Research Center. The first 
step is to add en route conflict- 
prediction capability to the CTAS 
software baseline. When fully 
developed, en route conflict- 
prediction and resolution capability 
should achieve many free-flight 
benefits without significantly altering 
current ATC procedures or requiring 
modification of airborne equipment. 

An initial field test was con- 
ducted from September 24 through 
October 8, 1996, at the Denver Air 
Route Traffic Control Center (ARTCC 
or ''Center"). The purpose was to 
conduct an initial evaluation of an 
en route conflict-prediction software, 
to evaluate an advanced tracker 
developed by the BDM Corporation 



Denver Center 
boundary 



En route 
traffic 



J94 





En route 
traffic 



Fig, 1 . Flight profiles for initial field test of en route conflict prediction. 



and the FAA Host computer tracker, 
and to get an initial look at opera- 
tional issues pertaining to en route 
conflict prediction. The Ames Lear 
Jet flew at angles of 0, 45, 90, 1 35, 
and 1 80 degrees with respect to jet 
route J94 near Scotts Bluff (see 
figure) to create (horizontal plane) 
conflicts with normal Center traffic, 
while maintaining normal safe 
altitude separation at all times. The 
NASA jet recorded differential 
Global Positioning System (GPS) 
position and velocity data to evalu- 
ate the trackers. A version of CTAS 
software, modified for en route 
conflict prediction, processed live 
radar, flight plan, and wind data for 
all aircraft in the Denver Center. 
Predictions of minimum horizontal 
separation between the NASA 
aircraft and each selected aircraft 
were begun nominally 20 minutes 



before the predicted time of mini- 
mum horizontal separation and 
continued just beyond minimum 
separation. 

A preliminary analysis of the 
data shows that minimum horizontal 
separation was predicted with 
1 .5-mile accuracy during the 
15-25 minute period before the 
time of actual minimum separation. 
This is well within the FAA separa- 
tion criteria of 5 miles for en route 
airspace and 3 miles for the terminal 
area. These results are based on 
horizontal separation data related to 
the distances between the NASA Jet 
and each of 59 other aircraft of 
various types over 8 days of test 
flights. 

Point of Contact: D. McNally 
(650) 604-5440 
dmcnally@nnail.arc.nasa.gov 



(i [. H A I, (> 1 \ [ I, A V I A T ION / Affordability 



23 



Assessing Controller 
Performance Under 
Simulated Free-Flight 
Conditions 

Roger Remington^ James Johnston, 
Eric Ruthruff, Maria Romera 

Traditional air-traffic control 
procedures restrict air traffic to 
designated routes and altitudes, 
analogous to highways in the sky, 
it Is well understood that these 
regulations, although promoting safe 
air travel, impose a price on the 
airlines in increased flight time and 
increased fuel consumption. NASA 
is presently working with the Federal 
Aviation Administration (FAA) to 
develop plans to relax these con- 
straints on air traffic so that airlines 
would be free to choose routes that 
minimized delays and fuel costs. 
Free-flight refers to a family of 
proposed concepts ranging from the 
relaxation of a few constraints to 
total freedom of choice of flightpath. 
The goal of this joint NASA-FAA 
activity, the Advanced Air Transpor- 
tation Technology program, is to 
ensure that the relaxation of con- 
straints on airspace operations do 
not compromise safety. The Cogni- 
tion Laboratory at Ames Research 
Center has been investigating the 
effects that relaxing airspace con- 
straints can have on controller 
performance. 

The extent to which constraints 
can be relaxed and free flight 
enhanced is assumed to depend, in 
part, on the workload of fllghtcrew 
and air-traffic control (ATC). Regard- 
less of how future responsibility is 



divided between pilots and control- 
lers, the safety of the overall system 
will depend heavily on the ability of 
controllers to perform their jobs 
well. One of the tasks of controllers 
will be to look at situation displays 
and understand the traffic, including 
the position and paths of the aircraft 



displayed. Under the present system 
this kind of ''situation awareness" is 
facilitated by virtue of traffic patterns 
that are orderly and have a familiar 
structure. Research on human 
cognition suggests that reducing the 
structure present in traffic flows 
might make it more difficult for 





Fig. 1. A traffic display, (a) With route restrictions, (b) Without route 
restrictions. 



24 



Global Civil A v i a t i o p^ / Affordability 



A E R S A IT 1 C S A S D S P A C E T R A S S P R T A T I S 
T E C H S L C Y E S T E R P R 1 S E 



40 




Routes 



12 





D Routes- 
alt restrictions 


1 


O Free 
■ Free- 
alt restrictions 


16 




Number of planes 





20 



Fig. 2. Conflict detection time as a function of number of aircraft on screen 
for the four restriction conditions. 



controllers to maintain situation 
awareness. 

A series of experiments has been 
conducted to determine empirically 
whether controller situation aware- 
ness is hindered by the loss of traffic 
constraints under free flight. To 
measure situation awareness, the 
experiments tested the speed with 
which controllers could find con- 
flicts in traffic displays. The experi- 
mental design contained four 
conditions: the status quo (both fixed 
routes and fixed altitude restrictions), 
free flight (neither fixed routes nor 
fixed altitude restrictions, fixed 
routes only, or fixed altitude restric- 
tions only. Part (a) of the first figure 
shows a traffic display with route 
restrictions; part (b) of the first figure 



shows a traffic display with the same 
number of aircraft in which route 
restrictions have been eliminated. 
In addition, the number of planes 
displayed was systematically varied. 
The second figure shows the results 
of one of the experiments in which 
experienced controllers detected the 
presence of conflicts with the aid of 
a computer conflict-assessment tool. 
The number of aircraft had strong 
effects on performance; detection 
times increased with increased 
number of planes in all conditions. 
Removing route restrictions did not 
negatively affect performance, but 
removing altitude restrictions did 
have a modest negative effect. 
Similar results have been obtained 
in conditions without computer- 



assisted conflict detection. The 
overall results to date provide no 
reasons for pessimism about the 
feasibility of considerable progress 
toward the objective of free flight. 

Point of Contact: R. Remington 
(650) 604-6243 
rremington@mail.arc.nasa.gov 



Evaluating Cockpit Display 
of Traffic Information 
Displays and Route 
Assessment Tools for a 
Free Flight Environment 

Vernol Battiste, Walter W. Johnson 

The Radio Technical Commis- 
sion for Aeronautics (RTCA) Task 
Force 3 report on free flight recom- 
mends the immediate initiation of 
''the development of standards for a 
cockpit situational awareness display 
of traffic information." In response 
to these recommendations, the 
Advanced Air Transportation 
Technologies Program and the Flight 
Management and Human Factors 
Division at Ames Research Center 
have initiated research on display 
designs and formats to support flight 
deck self-separation during free- 
flight operations. 

The primary goals of this effort 
are to (1 ) conduct research to 
support the development of display 
guidelines, and (2) evaluate and 
design cockpit display concepts and 
prototypes in support of cockpit- 
based separation during nominal 
and off-nominal operations from 
departure to arrival including the en 
route and transition phases of flight. 



CJ 1. H A 1, C 1 V 1 i. A \ [ A T ION / Affordahility 



25 



Three critical areas have been 
identified in the display design 
research: (1) alerting; (2) situational 
awareness; and (3) conflict resolu- 
tion. These three areas are interre- 
lated, and require an Integrated 
approach to arrive at good Human 
Centered Design recommendations. 
The research approach has been 
to identify attributes of displays that 
are critical to crew alerting and 
conflict resolution, and also to apply 
basic and advanced principles of 
good display design. Two critical 
attributes that have been identified 
are the surveillance data needed and 
the format of the display. The main 
principles guiding the Cockpit 
Display of Traffic Information (CDTI) 
design effort are to minimize head- 
down time and dwell time, to reduce 
scan frequency; and to improve 
information extraction. Another 
principle is minimizing the effect of 
alerts by providing enough advanced 
warning so pilots can have adequate 
time to plan for potential problems. 
Finally, the design considers that 
other cockpit duties prevent the 
crew from focusing their full atten- 
tion on the task of self-separation. 
Based on these principles and 
guidelines, an initial prototype 
Airborne Management of En route 
Separation (AMES) display was 
designed. To support the assessment 
of alternative route planning, a 
Route Assessment Tool (RAT) was 
designed and incorporated into the 
display design. The AMES display 
prototype was integrated with the 
B-747 Navigation display and the 
RAT controls were located on the 
top of the dash (see figure). The 
display was a two-dimensional 
track-up god's eye view of the traffic 
situation that provided pilots with an 




Fig. 1. Airborne Management of En Route Separation (AMES) display and 
Route Assessment Too! (RAT). 



evolving four stage (threat-level) 
depiction of the traffic, based on the 
output of a collision alert system 
developed at the Massachusetts 
Institute of Technology. In addition, 
this display allowed pilots to depict 
the predicted location of traffic up to 
1 minutes in the future, and to 
assess the safety of pilot-proposed 
route deviations. 

The AMES display and RAT 
were evaluated by 1 Boeing 747 



line pilots during a Crew Vehicle 
Systems Research Facility B-747 
simulation of free-flight operations. 
During the simulation the AMES 
display was compared with a more 
basic display that had only one level 
of alerting and no RAT tool. This 
simulation examined performance 
during eight en route traffic sce- 
narios incorporating multiple 
''traffic" aircraft, and one 'Intruder" 



26 



G I li A L C I V 1 L A V I A T ION / Affordcihility 



A t: R V A i T I f S A S D S P A C f T R A \ S P R T A T I V 
T E f H \ i G Y E S T t R P R 1 S E 



aircraft on a conflict course, and 
used flight procedures based, in part, 
on FAA rules guiding separation 
during VFR flight. The data from the 
simulation show that pilots using the 
AMES display were able to maintain 
a significantly wider margin of safety 
between ow^n-ship and intruder, and 
a lower overall level of conflict 
probabilities between own-ship and 
intruder, and between own-ship and 
other context aircraft. The analysis 
also shov\^ed no increase in the 
amount of maneuvering with the 
AMES compared with the basic 
display, and in fact showed less 
maneuvering with the AMES display 
under some circumstances. Crew 
comments on the display design 
were all very favorable, and sug- 
gested that self-separation would be 
possible if these types of displays 
and tools were available on the 
flight deck. 

Point of Contact: V. Battiste/ 

W, Johnson 

(650) 604-3666/3667 

vbattiste@mail.arc.nasa.gov 

wjohnson@mail.arc.nasa,gov 



when visibility is limited to '300 to 
700 feet by fog or rain. 

Two display technologies were 
investigated: a head-up display and 
an audio collision avoidance system. 
In the first, taxi guidance symbols 
are shown on a head-up display 
(HUD). A HUD is a clear, glass 
display mounted in the aircraft such 
that the pilot can see through it to 
view the out-the-window taxiway, 
and at the same time see information 
that is reflected on the display itself. 
This allow^s for display information 
viewing w^hile still maintaining out- 
the-window view^ing. 

The taxi symbology for the 
HUD is a subsystem of the Taxiway 
Navigation and Situation Awareness 
(T-NASA) System. The other sub- 
systems are a perspective moving- 
ma[) disf)lay, showing the plane's 
position at the airport, and three- 
dimensional i]-D] audio which 
alerts to the direction of potential 
traffic- incursions. For this project, 
superimposed HUD symbols 
showing current location and 
ground s|)eed, as well as pictorial 



augmentations to the scene, have 
been developed (see first figure). 
The superimposed text symbols, the 
"Inner, Alpha/Bravo" format, repre- 
sents the current runway or taxiway 
segment ("Inner Taxiway"), the last 
intersection passed ("Alpha"), and 
the next intersection upcoming 
("Bravo"). 

The pictorial scene augmenta- 
tions include visual enhancements 
that aid the pilot in following the 
taxiway clearance and completing 
turns. Using satellite positioning, the 
symbology is projected virtually in 
3-D such that the symbols move and 
change size in the same manner as if 
they actually existed in the out-the- 
window scene. We have termed this 
''scene- linked symbology." Vertical 
cones on the side of the commanded 
taxiway path depict the cleared 
route on the HUD in superimposed 
symbols. These are conformal and 
represent a virtual representation of 
the cleared taxi route on the HUD. 
The side cones and the centerline 
markings are shown repeated every 



20 GS 



Cockpit Displays for 
Low- Visibility Taxiing 

David Foyle, Elizabeth M. Wenzel, 
Durand R. Begault 

These cockpit display projects 
(as part of NASA's Terminal Area 
Productivity (TAP) program) are 
aimed at developing candidate 
cockpit display and audio technol- 
ogy for the improvement of civil 
transport (airline) taxi performance 
for navigation and traffic avoidance, 



Alpha Bravo 



* i A ' ^ 



Fig. /. Scene-finked symbols (in whiter) superimposed on runway scene. 



(i 1, !) r. A 1. C 1 V 



A \ I A T Ml N / Affurdahiiny 



11 



50 feet down the taxiway. The 
vertical development and constant 
spacing should yield increased 
capability for estimating ground 
speed, drift, and look-ahead capabil- 
ity for turns. Turn ''countdown'' 
warnings are shown in which each 
turn has countdown (4, 3, and 2) 
centerline lights that are 300, 200, 
and 100 feet, respectively, before 
each turn. This gives added distance 
cues for the turn. Virtual turn signs 
(with the arrows) give an added cue 
that a turn is necessary. In addition, 
the angle of the arrow on the sign 
represents the true angle of the 
turn (that is, 30 degrees right for a 
30-degree right turn). All of the HUD 
symbols are scene-linked (virtually 
projected as if they were real objects 
in the world), allowing the pilot to 
process the symbology in parallel 
with other traffic, including possible 
incursions. 

Simulation results have shown 
that the taxi HUD symbology allows 
pilots to taxi faster in low-visibility 
conditions and essentially eliminates 
straying off of the cleared ground 
taxi route. These benefits have the 
added bonus of also being associ- 
ated with decreased workload. 

The second display technology 
investigated was a 3-D audio 
Ground Collision Avoidance System 
(GCAS) and navigation system— 
spatially localized auditory traffic 
and navigation alerts. Additionally, 
the Crew Roles and Procedures 
project is assessing the effect of 
these advanced technologies on the 
cockpit cre?w and controllers, as well 
as recommending procedure modifi- 
cations as necessary. 

Broadly stated, the TAP initiative 
utilizes differential global position 



3-D audio warning for 
biunder avoidance 
under RVR 300 conditions 




GPS equipped 
obstacle 

(e.g., vehicle, 
other aircraft) 



3-D audio for 
directionalized 
warning signal 



^ 



Pilot hears 
virtual audio 
warning from 
the same angle 



Fig, 2. Illustration of the 3-D audio GCAS display for the terminal-area 
productivity program. 



sensing (DGPS) to track aircraft and 
other vehicles on the ground, and 
then uses this information to provide 
visual and auditory situational 
awareness by means of advanced 
guidance systems and displays. 
A 3-D auditory display is used as a 
ground collision avoidance system 
(audio GCAS; see second figure) 
to provide immediate situational 
awareness of crossing runways, of 
potential incursions, and of the 
aircraft being ^'off-track" (deviating 
from centerline). 

A full-mission simulation study 
was run using 1 2 crews in a 
B-747-400 simulator. It was hypoth- 
esized that there would be a signifi- 
cant pilot preference for an audio 
GCAS system to be included on the 
flight deck. That preference for the 
audio GCAS system was found. The 



main conclusions of the study were 
that an audio GCAS system would 
be useful for avoiding potential 
incursions under both normal and 
low-visibility conditions, and that an 
auditory system presenting incursion 
alerts would be a useful adjunct to 
a moving-map display. A positive 
preference was also found for the 
stereo headsets that were used to 
provide the 3-D audio cueing. 

Point of Contact: D. Foyle/ 

E. VVenzel 

(650) 604-3053/6290 

dfoyle@mail.arc.nasa.gov 

bwenzel@mail.arc.nasa.gov 



28 



G L B A I (^ ! V I L A V I A T ION / Affordahility 



A E R \ A i r I CS .4 S D S P A C E T R A S S P R T A T I S 
T E C H S I a \ E \ T E R P R I S E 



Multi-Sensor Image 
Registration 

Misha Pavel, Al Ahumada, 
Barbara Sweet 

The use of various imaging 
sensors that can penetrate obscuring 
visual phenomena (such as fog, 
snov^ and smoke) has been proposed 
to enable aircraft landings in low- 
visibility conditions. Images can 
be obtained from both active and 
passive sensors that are sensitive 
to electromagnetic energy in the 
visible, infrared, and millimeter 
wavelengths. Images can also be 
constructed from database informa- 
tion, by rendering an image of the 
database viewed from the estimated 
position of the aircraft. The focus of 
the current research effort is the 
development of image registration 
(superimposition) techniques in 
support of this application. 

Two challenges specific to 
multisensor registration must be 
overcome. First, the features in two 
different sensor images can be quite 
different. For example, the contrasts 
can be reversed; the runway might 
be bright in a visible-light image and 
dark in an infrared image. Previously 
derived procedures for image 
registration rely on a high degree of 
common features between the two 
images. FHowever, for this applica- 
tion, there is no guarantee that this 
exists; features apparent in one 
sensor might not be apparent in the 
other. Secondly, the two sensors are 
usually sited at different locations. 
This differential positioning can 
produce distortions of the images 




Fig. L Forward-looking infrared image (a), visible-light image (b), and the 
registered image pair (c). 



relative to each other which make 
registration more challenging. 

A standard method of registering 
two images is to find a spatial 
warping of one image that mini- 
mizes the sum of the squared 
differences between the two images. 
FHowever, for images from different 
sensors, this does not work because 
the images are not the same when 
the warping is correct. 

The figure shows results from 
NASA's new method. It estimates 
the image polarity relationship 
separately in local regions and drops 
regions from the difference measure 
that do not correlate well locally. 
The new method makes it possible 
to find the correct spatial transforma- 
tion despite local differences in 
feature polarity and despite image 
features that are unique to a sensor. 

Point of Contact: A. Ahumada/ 

B. Sweet 

(650) 604-6257/0006 

aahumada@mail.arc.na$a.gov 

bsweet@mail.arc.nasa.gov 



Flow Visualization of a 
Full-Scale Rotor in Hover 

Benton H. Lau, Alan J. Wadcock, 
Gloria K. Yamauchi 

The rotorcraft industry has 
strongly recommended that wake 
geometry and vortex strength be 
measured In addition to airloads 
and acoustic measurements during 
rotorcraft wind-tunnel tests. The 
existing aeroacoustic analyses 
require all three kinds of data for 
validation purposes, especially when 
predicting blade-vortex interaction 
(BVI) noise. Accounting for blade 
deflection is also important in 
accurately predicting BVI noise and 
in validating aeroelastic analyses. In 
response to industry, Ames Research 
Center initiated a test program to 
evaluate several flow-visualization 
and blade-displacement measure- 
ment techniques for a full-scale 
rotor. Three techniques were 
evaluated on an XV-15 tiltrotor in 
hover in the Ames 80- by 1 20-Foot 
Wind Tunnel. 



(i i. H A 1, C MIL A V I A T ION / AffordahiUfy 



29 



Flow patterns on a rotating 
blade can be visualized using 
fluorescent microtufts. Thousands 
of fluorescent microtufts were 
installed on the upper surface of a 
rotor blade. When illuminated by 
a powerful ultraviolet strobe, the 
blade motion was frozen and the 
microtufts displayed the flow 
pattern. The microtuft images were 
subsequently captured on film over a 



range of thrust conditions including 
rotor stall. 

Rotor-wake geometry can be 
measured using laser-sheet and flow 
visualization (see figure). A smoke 
dispenser was installed near the 
rotor tip, and an optics pod was 
placed below the rotor plane on the 
wind-tunnel floor. When fed with a 
laser beam via a fiber optic cable, 
the pod generated a thin laser sheet 




Fig. 1. XV-15 tiltrotor and laser sheet setup in the 80- by 12()-Foot Wind 
Tunnel at Ames. 



and illuminated the entrained smoke 
in the rotor wake. Images of the 
rotor-wake geometry were recorded 
on video at several thrust conditions. 
In addition to testing the standard 
tiltrotor blades, a series of subwings 
were tested and their effects on wake 
geometry and tip-vortex roll-up were 
evaluated. 

Blade displacement was mea- 
sured using a stereo imaging tech- 
nique. Retro-reflective targets were 
placed along the blade span on the 
upper surface. Two asynchronous 
video cameras and strobes mounted 
in the wind-tunnel ceiling were used 
to capture the frozen blade motion. 
The blade displacement or deforma- 
tion can be deduced from the 
positions of these targets. 

The success of these test tech- 
niques provides promising results for 
hover testing of a full-scale rotor. 
The test techniques can be applied 
to forward flight testing. 

Point of Contact: B. Lau 

(650)604-6714 

blau@mail.arc.nasa.gov 



Skin-Friction 
Measurements on a 
Hovering Rotor 

Alan J. Wadcock, Gloria K. Yamauchi 

A technique for acquiring skin- 
friction measurements on a fixed 
wing using a fringe-imaging skin- 
friction (FISF) technique was devel- 
oped at Ames Research Center in 
the early 1 99()s. As long as the flow 
conditions are steady, as they are in 
hover, this technique can also be 
applied to rotary wings. During a 



30 



(; L H A I C ! V 1 1, A V 1 A T i N / AfforciabHUy 



A i K U \ t / I f ( s 



,] \ D Space T r a \ s p o r r a t i o s 
T E ( n s n I (, \ E \ t e r p r i s e 



.010 


r- 












.008 


~ 




\ 








.006 


_ 




\ 








X 


- »- 






V 






.004 


- 


^^ 




V 






.002 


1 1 


\ 


1 1 


1 t 1 


1 1 1 


. 1 1 


u 




.2 .^ 


\ 

x/c 


.6 


.8 


1.0 



Fl^. L Chordwise skin-friction coefficient. 



hover test of a full-scale XV- 1 5 
tiltrotor in the 80- by 1 2n-Foot Wind 
Tunnel at Ames, the FISF technique 
was demonstrated for the first time 
on a rotor. 

The objective of this first 
demonstration was to determine the 
practicality of the FISF technique 
when applied to a full-scale rotor. 
Since the FISF technique had not 
been used previously in a rotating- 
flow environment, the effect of 
centrifugal force on the fringe 
patterns was a concern, as were the 
transient effects caused by starting 
and stopping the rotor. The effect of 
transients was minimized by remain- 
ing on the test condition for a 
suitable length of time. The effect of 
centrifugal force was shown to be 
negligible, once the oil-film thick- 
ness was reduced sufficiently. 
ChordwMse skin-friction coefficient 
measurements were acquired at one 
low-thrust condition at three blade 
radial stations. Measurements on the 
XV- 1 5 tiltrotor blade at a radial 



station of r/R = 0.24 are shown in 
the accompanying figure for a tip 
Mach number of 0.59, a thrust 
coefficient of 0.0035, and a 
Reynolds number of 1 .1 million. 
The measurements clearly indicate 
the location of the transition from 
laminar to turbulent flow. 

Point of Contact: A. Wadcock 
(650) 604-4573 
awadcock@nnail.arc.nasa.gov 



Rotor Data Correlation 

Randall Peterson 

The use of wind-tunnel test 
measurements, flight-test measure- 
ments, and analytical predictions 
plays a key role in the development 
of new rotor systems. Such tests are 
typically performed by using a range 
of rotor-system sizes and wind- 
tunnel test facilities. To ensure the 
accuracy of wind-tunnel testing 



methods, an experimental validation 
study was conducted using test 
results from model- and full-scale 
tests to compare with flight-test data. 

Accurate measurements ot 
rotorcraft performance as measured 
in a wind tunnel are strongly influ- 
enced by the test-section configura- 
tion, that is, whether it is a closed or 
open jet. The influence of wind- 
tunnel walls on the induced velocity 
of lifting bodies has been studied by 
many researchers over the years. 
Methods have l)een developed to 
adjust the angle of attack and 
dynamic pressure for fixed-wing 
aircraft in wind tunnels to approxi- 
mate free-flight conditions. These 
methods have largely been adopted 
by the rotorcraft community with 
very little testing to verify their 
applicability to helicopter rotors 
and flight-test measurements. 

Tests conducted by the 
Deutsche Forschungsanstalt fur 
Luft- und Raumfahrt e.V. (DLR) in 
the Duits-Nederlandse Wind Tunnel 
(DNW), under the auspices of the 
U.S. Army/German Memorandum 
of Understanding on Cooperative 
Research in the Field of FHelicopter 
Aeromechanics, have provided data 
suitable for evaluating these meth- 
ods. A 40"/) scale-model BO-105 
rotor was tested in five different 
wind-tunnel test sections: (1) 6- by 
6-m closed, (2) 8- by 6-m closed, 
(3) 8- by 6-m open slots, (4) 9.5- by 
9.5-m closed, and (5) the 8- by 6-m 
open jet. These data along with full- 
scale data from an Ames 40- by 
80-Foot Wind Tunnel test and a 
DLR flight-test program provided a 
means to evaluate wind-tunnel wall- 
correction methods specifically for 
helicopter rotors. Good correlation 



(; 1. 15 A 1. (] [ \ I I. Aviation / Affordahilily 



31 





45 knots 






.007 


r- 


— O — Experimental data 




D 




- 


- - n - - CAMRAD/JA — no wall correction 




/ 


.006 


- 


A — CAMRAD/JA — with wall correction 




/ 


.005 


- 




D / 


/ 


^•°" 


- 


* 

/ 

/ 






o 


- 


j^iX 






.003 


"~ 


a>^ 






.002 


- 


g,^:^^^©^''"*'^ 






.001 


- 


1 1 1 1 1 1 1 i 1 1 1 1 1 1 1 1 1 


1 1 1 


1 1 1 













.02 .04 .06 .08 


.10 


.12 



Fig. 1. Rotor power with and without angte-of-attack correction applied to 
the trim conditions of the rotor in the analysis. 



of rotor power over a range of 
advance ratios for these three 
datasets was shown using wall- 
correction methods after accounting 
for trim differences between the 
datasets. 

Recent rotor-data correlation 
work has focused on the ability of 
analytical prediction methods such 
as CAMRAD/JA (Comprehensive 
Analytical Model of Rotorcraft 
Aerodynamics and Dynamics, 
Johnson Aeronautics) to accurately 
predict rotorcraft performance. Full- 
scale BO-105 rotor-system data 
acquired in the Ames 40- by 80-Foot 
Wind Tunnel are being used in this 
analytical study. From previous 
experimental comparisons, it was 
shown that the rotor-performance 
measurements were strongly 



influenced by the presence of the 
wind-tunnel walls. Using the wall- 
correction method of Giauert, the 
angle of attack of the rotor system in 
the tunnel was corrected and used in 
the trim conditions of the CAMRAD/ 
JA analysis. As shown in the figure, 
the accuracy of predicted rotor 
power is significantly improved 
when the angle-of-attack correction 
of Glauert is applied to the trim 
conditions of the rotor in the 
analysis. The analytical predictions 
of rotor power as a function of rotor 
thrust using a free-wake analysis 
are shown. 

Point of Contact: R. Peterson 
(650) 604-5044 
rpeterson@maiLarc,nasa.gov 



Canard RotorAVing 
Hover Test 

Stephen Swansori/ John Madden 

In a joint program between 
NASA and McDonnell Douglas 
Helicopter Systems, a one-half-scale 
rotor model of the McDonnell 
Douglas Canard RotorAVing (CRW) 
rotorcraft was tested in the 80- by 
1 20-Foot Wind Tunnel at Ames 
Research Center. This test was 
conducted to evaluate the hover 
performance of the unique elliptical- 
airfoil rotor being planned for the 
CRW. The CRW is a stoppable-rotor 
design which can hover and fly at 
low-speeds like a conventional 
helicopter, whereas in its stopped- 
rotor mode it can fly at high speeds 
comparable to those of fixed-wing 
aircraft. Initial concepts include a 
land- or ship-based medium-range 
vertical takeoff and landing, 
remotely piloted vehicle. 

A two-bladed teetering rotor is 
used to generate the required lift for 
hover and low-speed forward flight. 
Once the rotorcraft is at a sufficient 
forward velocity, the required lift 
generation is transferred from the 
rotor to a canard and horizontal tail. 
What is unique about this rotor is 
that it Is driven by high-pressure 
air exhausted at the rotor tips as 
opposed to the conventional heli- 
copter in which the rotor is shaft- 
driven. In the flight vehicle, 
conventional turbofan engines are 
used to supply the high-pressure air 
to the rotor. Because the CRW's 
rotor is stopped to allow high-speed 
forward flight, the rotor's airfoil cross 
section must be elliptical. This is a 
compromise between the optimum 



32 



G ;, {) H A [, C I V I L A V I A T I N / Affordahility 



Aero .v ,4 i t 1 c s a s d S p a c t: T r a \ v i> r i a t i s 
T E C H s L y E s r E R P R I S E 




Fig. h Isolated rotor model installed in the Ames 80- by 120-Foot Wind Tunnel. 



airfoil shape for conventional rotor 
flight and that for high-speed 
stopped-rotor flight. 

The fixed-wing aerodynamic 
characteristics of the complete 
aircraft were evaluated in earlier 
tests. In this recent test, the isolated 
rotor was evaluated in hover in the 
Ames 80- by 1 20-Foot Wind Tunnel; 
that tunnel was used because it has 
the unique capability to provide 
heated, high-pressure air that could 
be used to simulate the exhaust 



gases of the CRW's engine. The 
figure shows the isolated rotor 
installed in the test section and 
connected to the high-pressure air 
system. Preliminary results indicate 
that the rotor was capable of gener- 
ating the required lift for takeoff and 
hovering flight. Comparisons with 
predicted aerodynamic performance 
showed that the actual thrust 
generated was greater than predicted 
for the same power input. On the 
other hand, comparisons of the 



actual and predicted internal 

pressure measurements indicated 
higher than expected duct losses 
within the rotor internal piping. 

Point of Contact: S. Swanson 
(650) 604-4565 
sswanson@mail.arc.nasa.gov 



G I i) B A I. (^ I V 1 1. A V [ A ! ION / Affordabilily 



33 



in-flight Dynamic Stall 
Research 

Robert Kufeld^ William Bousman 

The objective of this work Is to 
improve the complex dynamic-stall 
aerodynamic models of a helicopter 
rotor in high-speed, high-load 
conditions. The improvement of 
these models should serve to reduce 
the design and development cost 
of new helicopters, because the 
dynamic-stall condition is the 
driving force in determining the 
strength of the helicopter control 
system. Test conditions were identi- 
fied and studied from the flight- 
maneuver data collected during the 
UH-60A Airloads Program, which 
provides information about the 



dynamic-stall phenomena. UH-6()A 
airloads data are available to NASA's 
customers for similar studies through 
an electronic database and search 
engine called 'TRENDS/' which is 
maintained at Ames Research 
Center. 

The figure shows some of the 
complex loading a helicopter rotor 
is subjected to during a high-speed 
turn. The data show the measured 
pitching-moment calculation derived 
from pressure transducers located 
at radial station r/R = 0.92, at an 
advance ratio of 0.39 and a load 
factor of 1 .5. The figure clearly 
shows the five rapid and large 
negative pitching-moment depar- 
tures the blade experiences as it 
makes one revolution during this 




90 



135 180 225 

Blade azimuth (deg) 



270 



310 



360 



Fig. L Section moment versus rotor dzimuth for r/R = 0.920 in a high-speed 
turn; advance ratio = 0.390 and load factor = 1.50. 



test condition. The two pitching- 
moment departures near 280 and 
355 degrees have been clearly 
identified as dynamic-stall events. 
The two pitching-moment departures 
near 95 and 1 35 degrees have been 
linked to passage of shock waves 
over the airfoil; the final pitching- 
moment departure near 50 degrees 
appears to be caused by a combina- 
tion of two effects: supersonic 
aerodynamic and dynamic-stall 
vortex shedding. These data show 
the need for dynamic-stall model 
Improvement at both the traditional 
low-speed range of the fourth 
quadrant and the unexpected high- 
speed range found in the first 
quadrant of the rotor. 

Point of Contact: R. Kufeld 
(650) 604-5664 
rkufeld@mail.arc.nasa.gov 



Stall Control of Helicopter 
Rotors 

Khanh Q, Nguyen 

A control system based on 
blade-pitch actuation was developed 
to suppress stall on helicopter rotors. 
Stall suppression would allow the 
helicopter flight envelope to be 
expanded and would alleviate stall- 
induced loads and vibration. The 
control system was implemented 
in a comprehensive rotorcraft 
analysis, which was based on a 
finite-element method and which 
included unsteady aerodynamic 
effects (dynamic stall) and nonuni- 
form inflow models. The stall- 
suppression system is based on a 
transfer-matrix approach; it employs 



34 



(i i, H A I, C I \ I I. A V I A r i N / Affordabilily 



A E S \ A a T I € S A S D S P A C E T R A S S P R T A T I S 
T E C H S L C Y E S T E R P R I S E 



a Stall index as a measure of stall. 
The results showed that stall could 
effectively be suppressed using 
higher harmonic control at both 
cruise and high-speed flight condi- 
tions. The effectiveness of the stall- 
suppression system is shown in the 



figure in which the stall regions are 
shown without the system (part (a)) 
and with the system (part (b)) at a 
high-thrust, high-speed flight condi- 
tion. The control amplitude was 
small, less than 1 degree. 



(a) 




(b) 




180 
Azimuth (deg) 270 



360 1 



Fig. 7. Stall regions over rotor disk without the stall-suppression system (a) 
and with the system (b). 



In general, however, stall 
suppression did not guarantee 
performance improvements. The 
results also showed the distinction 
between stall suppression and 
performance improvement with 
active control. When the controller 
aimed to reduce the shaft torque, 
rotor-performance improvement 
could be achieved, but with a small 
degradation in stall behavior. 

Point of Contact: K. Nguyen 
(650) 604-5043 
knguyen@mail.arc.nasa.gov 



Apache AH-64D Flight 
Test Predictions 

Earl P. N. Duque 

Recent design changes to the 
AH-64D (Apache) helicopter have 
created unfavorable handling 
qualities that may be caused by 
changes in the aerodynamic flow 
field. Engineers at McDonnell 
Douglas Helicopter Systems needed 
to quickly understand the flow 
characteristics and to evaluate 
potential ways to alleviate the 
problem. Computational fluid 
dynamics (CFD) is an ideal tool for 
examining the effect of late-cycle 
design changes because it is much 
faster and cheaper than flight testing. 

The objective of this project was 
to use CFD methods to study the 
flow field around the helicopter in its 
current configuration and then with 
several variations which designers 
proposed to improve the handling 
qualities. The simulation results had 
to be obtained quickly in order to 



G I H A i. Civil Aviation / Affordabilil] 



35 



decide which designs would be built 
and tested in flight. 

Inviscid simulations were done 
with the TIGER code by John Melton 
of Ames Research Center (ARC). Its 
unstructured Cartesian formulation 
largely bypasses the time-consuming 
and tedious task of grid generation, 
allowing for fast turnaround of very 
complicated models. Results from 
the initial geometry (see figure (see 
Color Plate 3 in the Appendix)) 
influenced two of the proposed 
design solutions. Subsequent study 
of five new designs showed that one 
did not warrant further investigation. 
Since the validity of an inviscid 
approximation is questionable in a 
complex, vortex-dominated flow, 
intensive study of a geometrically 
simplified model (the isolated 
fuselage without engine, wing, or 
tail surfaces) using the thin-layer 
Navier-Stokes equations was con- 
ducted using the OVERFLOW code 
by Pieter Buning of ARC. These 
simulations gave further credence 
to the baseline inviscid computa- 
tions. The combination of complex- 
geometry/simplified-physics 
simulations with simplified- 
geometry/complex-physics simula- 
tions provided McDonnell Douglas's 
engineers with an overall under- 
standing of the Apache flow field 
and with quick feedback on pro- 
posed design changes. 

Point of Contact: E. Duque 
(650) 604-4489 
eduque@mail.arc.nasa.gov 



Rotor-Wake/Fuselage 
Interaction 

Paul M. Stremel 

The interaction of the wake 
generated by a rotor with the 
fuselage of the vehicle has profound 
effects on the overall efficiency of 
the vehicle. These effects have been 
clearly demonstrated in flight as 
vehicle vibration and download. 
Although they are clearly demon- 
strable in flight, the quantitative 
aspects of the interactions are not 
clearly understood. Numerous 
numerical and experimental investi- 
gations have been conducted in 
efforts to better understand the rotor- 
wake/fuselage interaction. However, 
each of these investigations concen- 
trated on a unique rotor/geometry 
combination for which conclusions 
can be drawn only for that particular 
configuration. A generalized method 
for optimizing rotor performance 
and minimizing fuselage interaction 
is not forthcoming. As a result, 
numerical methods must be applied 
to each configuration to better 
identify the aerodynamic effects 
of the rotor/fuselage interaction. 

The present research was 
conducted to obtain baseline 
interaction results — aerodynamic 
and acoustics data— for the flow 
surrounding the Tilt Rotor Aero- 
acoustic Model (TRAM) isolated- 
rotor configuration. It is extremely 
important to minimize the interac- 
tion of the support hardware with 
the rotor to insure quality data. 
However, because of geometric 
limitations of the support hardware, 
tunnel installation, and test enve- 
lope, a perfectly isolated rotor 
cannot be tested. The numerical 



method used to model the TRAM 
isolated -rotor configuration is a 
combination of the OVERFLOW 
computational fluid dynamics code 
and the PEGSUS domain connectiv- 
ity program. The isolated-rotor 
configuration consists of a rotor and 
a motor/gearbox assembly. In the 
current investigation, the rotor is 
represented by an actuator-disk 
model. Results were computed for 
Cj = 0.14 at angles of incidence of 
and 1 5 degrees. The results are 
shown in the first figure (see Color 
Plate 4 in the Appendix) for 0-degree 
incidence and in the second figure 
(see Color Plate 5 in the Appendix) 
for 15-degree incidence. Shown in 
each figure are the vertical velocity 
contours and the flow-field stream- 
lines. The vertical velocity contours 
are shown at the rotor plane and at 
a station on the model support. In 
comparing the streamline results, the 
rotor wake for the 0-degree angle of 
incidence directly interacts with the 
motor assembly. However, at the 
1 5-degree angle of incidence the 
rotor wake is completely clear of the 
motor assembly. In comparing the 
velocity contours at the rotor plane, 
the contours indicate an interaction 
over the motor assembly for the 
0-degree angle of incidence but 
considerably less interaction at the 
15-degree angle. 

Point of Contact: P. Stremel 
(650) 604-4563 
pstremel@mail.arc.nasa.gov 



36 



G 1. B A I- 



C i V 1 L A V I A T ION / Affordability 



A E R \ A I r I c s A s D S p A c E Trassportatios 

T E C H S L G Y E S T E R P R I S E 



Navier-Stokes Simulation 
of High-Lift Aerodynamics 

Karlin Roth, Stuart Rogers 

The use of the methods of 
computational fluid dynamics (CFD) 
in order to understand the flow over 
high-lift wings configured for takeoff 
and landing with leading-edge slats 
and slotted tralling-edge flaps will 
contribute to the design of simpler or 
more efficient high-lift systems for 
subsonic transport aircraft. The long- 
range objectives of this work are to 
develop an accurate and efficient 
Navier-Stokes analysis method for 
flow over the high-lift systems of 
subsonic transport aircraft and to 
examine critical physics affecting 
high-lift system design. The particu- 



lar near-term objective is to predict 
the flow about a subsonic transport 
configured for high lift and to assess 
wind-tunnel interference effects on 
the flow about this test article. 

The OVERFLOW code was 
used to investigate the flow about a 
subsonic transport configuration. 
The simulations are steady-state, 
employ the Spalart-Allmaras turbu- 
lence model, and utilize multigrid 
convergence improvement schemes. 
The structured, overset grid system 
which discretizes the fuselage, high- 
lift wing elements, and nacelles 
consists of 1 6 million points within 
36 zones (see figure). During FY96, 
surface grids were generated from 
the computer-aided design defini- 
tion; volume grid methods including 



-vi-i 1 



■X- -rv■^\\■\■\\i-|■ 
fm 



-iJ 










element edge grids, wake place- 
ment, and wake smoothing strategies 
were developed, verified, and 
applied. Importantly, the simulation 
team demonstrated first-of-a-kind 
computations in utilizing a viscous 
CFD analysis for a complex three- 
dimensional high-lift configuration. 
The solution matrix includes a total 
of 1 8 cases which document the 
flow for a range of angles of attack, 
with and without powered engines, 
and in free-air and within the 
1 2-Foot Pressure Wind Tunnel at 
Ames Research Center. 

Comparison of the free-air and 
wind-tunnel simulations shows 
significant wall-interference effects 
which vary with the model attitude. 
In general, simulations conducted in 
the wind tunnel exhibit earlier stall 
than do the free-air calculations. 
Further, in addition to an earlier stall 
angle, powered simulations in the 
wind tunnel attain lift values near 
stall which are about 8% higher than 
the free-air values. In order to test 
the Wall Interference Correction 
System developed for the 12-Foot 
Pressure Wind Tunnel on a pow- 
ered, high-lift model, it was applied 
to the simulated data at selected 
conditions. The initial analysis 
indicates that this correction method 
may not be appropriate at conditions 
near maximum lift. 

Point of Contact: K. Roth/S. Rogers 
(650) 604-6678/4481 
kroth@mail.arc.nasa.gov 
srogers@mail.arc.nasa.gov 



Fig, 7. Overset grid system within cross section of a multielement 
high-lift wing. 



Global Civil Aviation / Affordability 



37 



Lift-jet Effects on 
Powered-Lift STOVL 
Model 

Karlin Roth 

During the transition from hover 
to wingborne flight, short takeoff and 
vertical landing (STOVL) aircraft rely 
on the direct thrust of lift jets to 
supplement aerodynamic v^ing lift. 
The jets usually induce a loss of lift 
and a change in the nose-up pitch- 
ing moment on the vehicle; both of 
these effects increase v^ith increasing 
forv^ard velocity. Thus, proper 
design of STOVL aircraft requires 
an understanding of the physics of 
the lift-jet/airframe interaction. 

The ability of three-dimensional, 
laminar and turbulent Navier-Stokes 
computations to simulate the 
performance of a geometrically 
simplified STOVL model during 
transition flight was investigated. 
Simulations were made using the 
OVERFLOW code for the model 
which consists of a 60-degree 
cropped delta wing planform; a 
simple fuselage shape blended to the 
wing; and tandem, circular, high- 
pressure-air lift jets that exit perpen- 
dicularly to the flat lower surface. 
Predictions were compared with 
measurements at a free-stream Mach 
number of 0.146, a 10-degree angle 
of attack, and with sonic lift jets. 
Flow visualization shows the 
structure of the wing leading-edge 
vortices and the lift jets; the deflec- 
tion of the lift jets is illustrated in the 
figure using Mach number contours 
in the model's symmetry plane. 

Although grid refinements 
improve the computational results, 
none of the computations com- 
pletely characterize the wing 




Fig. /. Mach number contours taken in the symmetry plane to illustrate the 
deflection of lift jets. 



leading-edge flow field or the flow 
within 2 diameters of the jet exits. 
For laminar calculations, the pres- 
sure suction peaks at the wing 
leading edge tend to be initiated 
farther inboard, to cover more area 
on the wing, to have slightly less 
suction, and to agree better with the 
measurements than the turbulent 
predictions. Notably, the simulations 
predict a jet-induced lift loss for 
the model that is mostly due to 
decreased suction at the wing 
leading edge. 

Point of Contact: K. Roth 
(650) 604-6678 
kroth@mail.arc.nasa.gov 



Vortex Core Detection 
for Computational Grid 
Refinement 

David Kenwright 

Predicting rotor wake is one of 
the most challenging problems in 
rotorcraft computational fluid 
dynamics (CFD). The rotor wake is 
the disturbed flow that is left behind 
as the rotor blade cuts through the 
air. The wake rolls up into vortices 
near the blade tips, a result of the 
pressure differences caused by the 
moving rotor. These vortices, an 
unwanted side effect of lift, cause 
control buffeting, airframe vibration, 
and noise. The requirement for low 
noise is particularly important for 



38 



G L B A I. C I V 1 L A V I A T I N / AffordahiHty 



A £ R \ A tl T I C S A S B S P A C E TraSSPORTATIOS 
T E C H S L C Y E S T E R P R I S E 



civilian helicopters and tiltrotors 
which operate in highly populated 
areas. 

A major problem with the CFD 
simulation of rotorcraft is that the tip 
vortices diffuse too quickly because 
of inadequate grid resolution. 
Adaptive grid-refinement schemes 
developed at Ames Research Center 
have been partially successful in 
capturing these vortices, although 
inadequate grid resolution near the 
vortex core causes discrepancies 
between computed and wind-tunnel 
results. 

A vortex-core detection tech- 
nique was implemented in the 
Unsteady Flow Analysis Toolkit 
(UFAT) to permit scientists to locate 
and visualize the regions of swirling 
or rotating flow. This technique is 
based on an eigenvector method 
developed at the Massachusetts 
Institute of Technology with funding 
from Ames. The figure (see Color 
Plate 6 in the Appendix) shows 
vortex cores extracted from a 
simulation of a helicopter rotor in 
forward flight. The vortex cores (red 
lines) become disjointed where the 
computational grid is too coarse. In 
this simulation, the grid was refined 
based on the vorticity magnitude. 
From the visualization, it is apparent 
that the grid refinement process 
failed to capture the vortex core. 

The vortex cores are extracted in 
a batch computation which can take 
from seconds to hours to complete 
depending on the size of the dataset 
and the complexity of the flow. For 
the rotorcraft simulation shown in 
the figure which consisted of 
800,000 tetrahedral elements, the 
extraction process took less than 
8 seconds on a Silicon Graphics 



Power Onyx workstation. The same 
technique has also been applied to 
much larger, unsteady simulations of 
aircraft to reveal the time-evolution 
of vortex cores. 

Point of Contact: D. Kenwright 

(650)604-1704 

davidk@nas.nasa.gov 



Wingtip Vortex Flows 

Jennifer Dacles-Mariani, 
Dochan Kwak 



The wingtip vortex flow has 
been the subject of numerous 
studies. Its significance has been 
seen in practical problems such as 
landing separation distances for 
aircraft, blade-vortex interaction, 
and cavitation on ship propeller 
blades (see first figure). Despite these 
numerous studies, most of the wake- 
vortex flow behavior is still not 
clearly understood. 



This work is part of an ongoing 
computational study of the tip-vortex 
flow through initial formation/ 
roll-up, growth, and decay. In this 
particular study, the near- and 
intermediate-fields behind the wings 
are investigated to (1) explore a 
more physical modeling of the flow 
by including the viscous upstream 
effects; (2) provide an improved 
vortex flow field to be used as inflow 
for the intermediate-wake; and 
(3) investigate the validity and 
applicability of a simplified Navier- 
Stokes solver using the velocity/ 
vorticity formulation in tip-vortex 
flow prediction. 

The approach taken is to derive 
a wake model from the full Navier- 
Stokes equations using the velocity/ 
vorticity formulation. It has the 
advantage of being able to retain 
some of the flow physics without a 
lot of empiricism. These equations 
allow physical diffusion in the 
streamwise direction, including 
viscous upstream effects and vortex 




Fig. 1 . Schematic of a downstream wake of an aircraft. 



G L B A L C I V I L Aviation / Affordabiliiy 



39 



stretching. For vortex-dominated 
flows, vorticity is the natural choice 
of variable. The accuracy of the 
method may be better preserved 
since it is being calculated directly. 

From this study, a detailed 
account of tip-vortex formation and 
of the roll-up process was accom- 
plished. The roles of numerics and 
turbulence models in tip-vortex 
prediction was also studied in detail. 
The preliminary results indicated 
that the velocity/vorticity formulation 
using the preconditioned conjugate 
gradient method was successful in 
predicting wake-vortex flow. Com- 
parison with measured data of a 
0.03 scale model of the Boeing 747 
aircraft, taken in the 80- by 120-Foot 



Wind Tunnel at Ames Research 
Center showed a similar trend 
between the measured and the 
computed upwash and downwash 
velocities (see second figure). The 
slopes of the velocity across the 
vortex were predicted well by the 
computation. Most of the disagree- 
ment is in the region where the 
sampling period of the measured 
data was not long enough, which 
resulted in the irregularity of 
the data. 

Point of Contact: J. Dacles-Mariani 
(650) 604-5369 
jdacles-mariani@mail.arc.nasa.gov 



.2 1- 



^ 



o 

o 



*-.1 



-.2 



-.3 



-M — Time averaged value (measured) 

Full Navler-Stokes formutation 

Modem 




-1.2 



-.8 -.4 




y/b 



.8 



1.2 



Fig. 2. Upwash and downwash velocity profile at x/c- 162 feet. 



Transonic Overset 
Potential Solver 

Terry Hoist 

The "full-potential+boundary- 
layer'' approach for transonic cruise 
analysis and design can be up to 
1 00 times faster than Navier-Stokes 
approaches and just as applicable 
over a wide range of cases involving 
weak shock waves with little or no 
boundary-layer separation. A fast 
full-potential capability using a 
zonal grid approach for general 
geometry modeling will be a valu- 
able tool for near real-time assess- 
ment of aerodynamic shapes in the 
design or wind-tunnel environments. 

The objective of the present 
work is to develop a full-potential 
analysis and design capability 
utilizing the Chimera zonal grid 
approach that will be compatible 
with the well-established 
OVERFLOW Euler/Navier-Stokes 
flow solver. Such an approach 
would be extremely cost effective, 
especially considering that all of 
these approaches would utilize the 
same problem setup and post- 
processing software, and to a large 
extent, the same grid-generation 
software. 

A transonic three-zone wing- 
fuselage solver (for inviscid flow) 
has been developed and validated 
using experimental pressure mea- 
surements (see figure). This half- 
million-grid-point computation, 
which includes surface definition, 
surface-grid generation, volume-grid 
generation, Chimera hole-cutting, 
donor-cell search, flow-field genera- 
tion, and PLOT3D file generation, 
required about 1 minute of computer 



40 



Global Civil Aviation / Affordabiiity 



Aefosa cries a ^ d Space Trassportatios 
Techsology E s t e r p r I s e 



1.0 I- 



? 

c 

0) 

o 

8 -.5 

9> 



CO 

t/i 

0) 



-1.0 




-1.5 
-.2 



o 
n 

A 

J I I I I L_ 



Comp(Y/8=0.17) 
Comp (Y/B=0.40) 
Comp (Y/B=0.85) 
Exp(Y/B=:0.17) 
Exp (Y/B=0.40) 
Exp (Y/B=0.85) 



± 



J I I I I I L. 



.2 .4 .6 

Normalized axial distance (x/c) 



(a) 




(b) 



.2 .4 .6 .8 

Normalized axial distance along fuselage (x/l) 



Fig. 1. Wing/fuselage surface-pressure comparisons RAE wing-fuselage, 
M^ = 0.8, a = 2°. (a) Upper wing; (b) upper fuselage. 



time on a single processor Cray C90 
computer. 

In developing this solver, a new, 
approximate Chimera donor-celi 
search algorithm has been devel- 
oped that is up to 100 times faster 
than previous techniques. For 
example, in two-zone cases, the 
new algorithm can find donor cells 
at a rate in excess of 60,000 cells 
per second on a single processor 
Cray C90 computer. This search 
algorithm has an approximate 
characteristic that will automatically 
utilize a nearest-neighbor cell when 
the actual donor cell cannot be 
found, thus reducing the chance for 
orphan points to exist. 

Point of Contact: T. Hoist 
(650) 604-6032 
tholst@mail.arc.nasa.gov 



Real-Time Particle Tracing 
in Time-Varying Flows 

David Kenwright^ David Kao 

Particle tracing is a scientific 
visualization technique that is 
widely used to study aeronautical 
simulations. In time-varying flows, 
particle tracing can be used to 
produce streaklines (a visualization 
technique commonly used in wind- 
tunnel experiments) by simulating 
the continuous injection of particles 
into a flow. To detect important flow 
features, several thousand particles 
must be injected and traced through 
a flow field by using numerical 
integration, which makes particle 
tracing computationally intensive. 
Consequently, streaklines are usually 
pre-computed in a batch process 



G L K A L (1 I V [ I, A V I A T ION / Affordabilitv 



41 



by using programs, such as the 
Unsteady Flow Analysis Toolkit, 
and then playing them back later in 
animations or videos — a process 
which can take hours. 

A real-time algorithm for 
computing and displaying streaklines 
known as the tetrahedra method, 
was developed by members of the 
visualization technologies group 
at Ames Research Center. The 
improved performance offered by 
the tetrahedra method now allows 
streaklines to be computed and 
visualized in real time on high-end 
Silicon Graphics, Inc. workstations. 
The benefit to scientists is immediate 
visualization of their data and the 
capability to interactively explore 
time-varying flows. Comparisons 
with conventional particle-tracing 
algorithms have verified that the 
tetrahedra mtHhod is accurate as 
well as fast. 

The tetrahedra method has been 
installed in two visualization systems 
developed at Ames: the Virtual 
Windtunnel (see the figure (see 
Color Plate 7 in the Appendix)) and 
the Unsteady Flow Analysis Toolkit. 

Point of Contact: D. Kenwright 

(650)604-1704 

davidk@nas.nasa.gov 



NASA Metacenter 

Mary Hultquist 

The NASA Metacenter is an 
exploratory project of the Parallel 
Systems Groups at Ames Research 
Center and Langley Research Center. 
The focus of the project is to achieve 
more effective use of NASA super- 
computers by making the systems 



more readily available to research- 
ers, and by providing quicker 
turnaround for batch jobs, a larger 
range of available resources for 
computation, and better distribution 
of the computational workload 
across multiple supercomputers. 

The current homogeneous 
configuration of the Metacenter links 
the IBM SP2 computers at Ames and 
those at Langley. With the systems 
similarly configured, each Center 
can rely on help from the other for 
troubleshooting difficult problems. 
Planning and problems are discussed 
at weekly teleconferences, and 
problems are resolved more quickly 
because the experience of personnel 
at both sites is utilized. 



The implementation of the 
Metacenter includes a ''peer sched- 
uler" to interact with the Portable 
Batch System (PBS) queuing system 
used on each system. In this model, 
as long as all systems have work to 
do, they act as independent comput- 
ers. But when the utilization of one 
system drops below a predefined 
utilization threshold, then peer 
scheduling begins. A system that 
is short of jobs, ''asks" the other 
systems in the Metacenter (its 
"peers'') for a list of all the queued 
jobs on the remote systems. It then 
goes through a series of checks, 
verifying that it can indeed run a 
specific job. If all the tests are 
passed, the scheduler requests the 





NASA Metacenter FY96 SP2 Utilization 




Four week floating average, Oct 1, 1995 - Oct 1, 1996 


100 


"_ 


— Target 




- 


— NASSP2 


80 




-- LaRCSP2 ^^ 




r^ V 


c 




/ \ 


o 




/ \ 


1 6° 


\ 


/ V^^ 


s 




r^ ^'N/ 


o 


- \ 


/\ /"^ 


? 


\ 


/ \f * ^ 


2 


\ 


/ -. ^ ' > . 


0) 40 


- \ 


/ ' \ -' ^ '^ 


-s 


\ 


/ ' \ ' * ' 


5 

1 


- \ 


/ OS upgrage ' ^ ' - \ ' 
I /^(AIX + PSSP) 


20 


~~ 


\a / ^"" 




' /. 


"^^ ** 


A 


1 ' 


""^.' 1 1 


Oct 95 


Apr 6 Oct 96 




Week beginning 



Fig. 1. FY96 utilization on tiie IBM SP2. 



42 



Global Civil Aviation / Affordability 



A E RO S A IT I C S A \ D S PAC E T R A S S F R T A Tl S 
T E C H S L G y E S T E R P R I S E 



PBS to move the job to the local 
server. The job is then run locally, 
and the results sent back to the user. 

This innplementation has 
resulted in a more balanced utiliza- 
tion of the Metacenter systems, as 
shov^n in the figure. The users are 
thus able to get jobs returned more 
quickly, because a job does not 
v^ait in a queue at one system while 
another system sits idle. Moreover, 
system downtime is decreased 
because personnel at both sites can 
work together to get down systems 
back into production. 

Point of Contact: M. Hultquist 

(650)604-0814 

maryh@nas.nasa.gov 



Portable Batch System 

David Tweten 

As of this year, all super- 
computers at Ames Research Center, 
whether production or testbed, and 
regardless of manufacture, run the 
same batch and cluster control 
system, the Portable Batch System 
(PBS). Most recently, Ames's two 
Cray C-90s, called the vonNeumann 
and Eagle, as well as the Center's 
Cray J-90 cluster, called the Newton, 
were converted to use the PBS as 
their sole production batch system. 
The PBS is a research batch process- 
ing and cluster control software 
package written at the Center. 

VonNeumann is NASA's largest 
supercomputer; it is used to perform 
cutting-edge research for NASA's 
aeronautics-focused programs and 
for base research and technology. 



Eagle is the production super- 
computer for general science and 
engineering research for the Office 
of Aeronautics and Space Transpor- 
tation Technology, Newton is a 
testbed supercomputer cluster, 
intended to ease migration of 
computer codes between vector 
computing and distributed- 
computing paradigms. The PBS 
allows construction of custom 
schedulers using any of three 
different languages. It is a key 
component in NASA's plan to 
develop metacenters, virtual com- 
puter centers composed of machines 
potentially supplied from several 
vendors, and potentially sited in 
several locations. 

The PBS beta program ended 
the fiscal year with 51 client sites, 
distributed among government, 
universities, and private industry. 

Point of Contact: D. Tweten 

(650)604-4416 

tweten@nas.nasa.gov 



Developing a Cluster 
Computer from 
Workstations 

Reese L. Sorenson 

Finding adequate computational 
resources to meet the ever-expanding 
needs of aerodynamic simulation by 
computational fluid dynamics (CFD) 
has always been a challenge, one 
exacerbated by recent budgetary 
constraints and redirected priorities. 
Facing a possible decrease of 
available computer time in the face 



of growing needs, it is important that 
a new source of computing power 
be found for researchers at Ames 
Research Center. 

It was realized that in terms of 
raw computer power the Silicon 
Graphics, Inc. (SGI) workstations 
sitting on the researchers' desks, 
considered together, had power 
approximately equal to that of a 
four-processor CRAY C-90. Since 
this resource was sitting idle every 
night and most weekends, a project 
was undertaken to link these work- 
stations, thus creating a cluster 
computer. 

Two needs had to be met to 
make this a reality. One was for fast 
interprocessor communication. A 
Grand Junction fast Ethernet data 
switch was obtained, and the 
24 most powerful of the approxi- 
mately 60 workstations in use were 
installed on the data switch. It is 
expected that this switch will at least 
double the interprocessor communi- 
cation speed for these workstations 
by eliminating ''data collisions" at a 
local network router. 

The second need was for 
scheduling software to queue jobs, 
oversee their execution, and gener- 
ally make the workstations function 
together as a computer. The Portable 
Batch System (PBS) software, 
developed locally at Ames and 
already functioning as the operating 
system on a CRAY supercomputer 
and on a dedicated workstation 
cluster, was ported to these 
machines and customized for this 
implementation. 

The cluster is scheduled to be 
available 1 2 hours per day through- 
out the year. Test cases have been 
run successfully on it. Recently 



^' •' *^ li A I, Civil Aviation / Affordability 



43 



acquired workstations, even more 
powerful than those originally used, 
are being installed in the cluster. 
Documentation to be posted on the 
World Wide Web has been written. 
The significance of this project 
is twofold: first, the critically impor- 
tant work of aerodynamic simulation 
at Ames will continue to have 
adequate computational resources; 
second, the software techniques and 
the experience of this project are 
available to others facing similar 
constraints. 

Point of Contact: R. Sorenson 
(650) 604-4471 
rsorenson@mail.arc.nasa.gov 



Planar Doppler 
Velocimetry Using 
Pulsed Lasers 

Robert L. McKenzie 

Recently, planar doppler 
velocimetry (PDV) has been shown 
by several laboratories to offer an 
attractive means for measuring three- 
dimensional velocity vectors every- 
where in a light sheet placed in a 
flow. Unlike other optical means of 
measuring flow velocities, PDV is 
particularly attractive for use in large 
wind tunnels where distances to the 
sample region may be several 
meters, because it does not require 
the spatial resolution and tracking of 
individual scattering-particles or the 
alignment of crossed beams at large 
distances. To date, demonstrations 
of PDV (called ^'doppler global 
velocimetry" by some authors) have 
been made either in low-speed flows 



without quantitative comparison 
with other measurements, or in 
supersonic flows where the Doppler 
shift is large and its measurement is 
relatively insensitive to instrumental 
errors. Moreover, most reported 
applications have relied on the use 
of continuous-wave lasers, which 
limit the measurement to time- 
averaged velocity fields. 

The objective of this study has 
been to quantitatively determine the 
limits of PDV capabilities for appli- 
cations in large-scale wind tunnels 
that are intended primarily for 
production testing of subsonic 
aircraft. For such applications, the 
adequate resolution of low-speed 
flow fields requires accurate mea- 
surements of small Doppler shifts 
that are obtained at distances of 
several meters from the sample 
region. In addition, the use of 
pulsed lasers provides the unique 
capability to obtain not only time- 
averaged fields, but also their 
statistical fluctuation amplitudes and 
the spatial excursions of unsteady 
flow regions such as wakes and 
separations. 

To accomplish the objectives, 
the PDV measurement process was 
first modeled and its performance 
evaluated computationally. The 
noise sources considered included 
those related to the optical and 
electronic properties of charge- 
coupled device (CCD) arrays and 
to speckle effects associated with 
coherent illumination from pulsed 
lasers. The signal-noise estimates 
were incorporated into the PDV 
signal analysis process and com- 
bined with a spectroscopic model 
of the iodine vapor cell used to 
discriminate Doppler frequency 



shifts and with computed scattering 
signals using a Mie scattering theory 
for polydisperse smoke particles. The 
relevant parameters incorporated a 
range of practical aerodynamic test 
conditions and facility sizes. The 
results (1) helped define the opti- 
mum values of the instrument 
parameters, (2) showed that the 
expected signal levels from a 
practical PDV system were suffi- 
ciently large to allow its useful 
application in large facilities, and 
(3) showed that the expected 
velocity measurement uncertainties 
were small relative to the mean 
velocities of interest for most sub- 
sonic, large-scale wind-tunnel 
testing. 

PDV performance was then 
demonstrated by using several 
bench-top setups, including a 
rotating wheel where the exact 
velocity field was always known 
everywhere, a turbulent air jet with 
flow properties that were calibrated 
using pitot and hot-wire probes, 
and stationary solid and aerosol 
targets where only the PDV noise is 
observed. Images from these targets 
were used to validate the model 
description of noise magnitudes and 
to confirm the minimum velocities 
that could be resolved. The bench- 
top experiments also allowed the 
optical configuration and calibration 
procedures of the instrument to be 
developed into a robust and easily 
operated system that is applicable to 
the environment of large-scale wind 
tunnels. An example of some results 
is shown in the figure (see Color 
Plate 8 in the Appendix) where the 
average axial velocity field and its 
root-mean-square fluctuation 
amplitudes are displayed for the flow 
in a low-speed, turbulent, free jet. 



44 



G L B A I C I V 1 L A V 1 A T ION / AffordabiUty 



A E R S A i r f C S A S D S P A C E T R A \ S P R T A T I V 

Techsoiogy E s r e r p r I s e 



The plane of the image is normal to 
the jet centerline. The velocity 
profiles are taken along the grid lines 
shown in each image. The velocity 
field/ part (a) in the figure, shovi^s the 
average velocity distribution of the 
core flow and the fluctuation field; 
part (b) of the figure shows the high 
relative fluctuation amplitudes in the 
turbulent shear layer at the edge of 
the jet. These measurements were 
made with a preliminary PDV 
optical system that was capable of 
determining only one velocity 
component. The combined use of 
two additional systems would allow 
the determination of all three 
velocity components simultaneously 
from each laser pulse. 

The PDV measurement capabili- 
ties developed in this study demon- 
strate that PDV offers significant 
advantages when compared with 
other means of measuring velocity 
fields in large-scale, subsonic wind 
tunnels. Measurements obtained 
from each laser pulse allow the 
accurate resolution of complete 
velocity vector fields with resolu- 
tions of instantaneous speeds as low 
as 2 m/sec. Much smaller average 
speeds can also be resolved. From 
the pulse-to-pulse data, both mean 
velocities and their fluctuation 
amplitudes can be determined in 
any aerosol-seeded flow with 
adequate optical access, including 
complex, three-dimensional, turbu- 
lent flows. The method has no upper 
limit above which velocity can not 
be measured, other than in extreme 
hypersonic flow situations contain- 
ing strong shock waves; in the latter 
case, the necessary scattering 
particles or aerosols in the flow do 



not follow the sudden changes in 
flow speeds or directions. The 
method has further advantages: 
the measurement procedures are 
compatible with the production 
testing environment of large-scale 
wind tunnels, the experimental setup 
is relatively simple, and the system 
could be operated by trained wind- 
tunnel technicians as a production 
instrument. 

Point of Contact: R, McKenzie 

(650)604-6158 

rmckenzie@mail.arc.na$a.gov 



Pressure-Sensitive Paint 
and Photogrammetry for 
Aeroelastic Experiments 

Edward T. Schairer, 
Lawrence A. Hand 

Methods for predicting the 
aeroelastic stability of aerospace 
structures are generally based on 
linearized aerodynamic theory. The 
accuracy of these methods suffers 
when nonlinear flow features (for 
example, shock waves, vortices, 
separated flow) interact with the 
aircraft structure. For example, at 
transonic speeds, where significant 
shock waves may be present, two- 
and three-dimensional wings tend to 
flutter at lower dynamic pressures 
than predicted by linear flutter 
analyses. 

Since the late 1 970s, consider- 
able progress has been made in 
using computational fluid dynamics 
(CFD) to account for nonlinear 
aerodynamics in aeroelastic analy- 



ses. Before these more powerful 
methods can be of practical value to 
aircraft designers, however, they 
must be tested by comparing their 
results with reliable experimental 
data. In particular, it is important to 
verify that unsteady, nonlinear 
aerodynamic features and their 
interactions with the structure are 
properly represented. 

To help address this need, 
simple, small-scale, aeroelastic 
experiments were conducted in the 
High Reynolds Number Channel 2 
(HRC-2) at Ames Research Center. 
The objective of the experiments 
was to (1 ) evaluate the use of 
pressure-sensitive paint (PSP) to 
measure unsteady pressure distribu- 
tions on aeroelastic models and 
(2) stereophotogrammetry to mea- 
sure unsteady model deformations. 
Semispan, clipped delta-wing 
models built from thin flat plates 
were tested at transonic speeds. 
The models were imaged by three 
synchronized black-and-white video 
cameras, and the data were recorded 
on videotape. Correlated images 
from two of these cameras were 
used to estimate the space coordi- 
nates of reference marks on the 
model at each instant, and the 
motion of each mark was estimated 
by fitting the deflection data with a 
damped sinusoid. The third camera 
was used for PSP. 

All the procedures needed to 
measure unsteady pressure distribu- 
tions using PSP were successfully 
demonstrated. The PSP measure- 
ments were seriously compromised, 
however, because the response time 
of the pressure paint was too slow, 
resulting in significant attenuation 



(j 1. H A 1. (^ i V i [. A V I A T [ {> \" / Af ford ability 



45 




(a) 



1.0 


r- 










- 




A 






.8 


- 




\!\M 


\f\/\y\^^ 




.6 
A 


- 


4 


> 


\j \/ ^^ ^'^ 

> J Ad 


Steady 

(final) 
Unsteady M\ 


.2 


1 


1 t t 1 


1 1 t 


1 1 1 1 1 1 1 1 1 i 1 






( 


) 


.2 


A 


.6 .8 1 


.0 








y/b 



(b) 



of the unsteady component of the 
PSP signal. The first figure presents 
typical PSP data from a run 
(Mach = 0.85), where the wingtip 
was deflected and released. The gray 
levels in the image (part (a)) which 
was acquired just after the wingtip 
was released, are proportional to the 
pressure coefficient. Part (b) of the 
figure compares the pressure coeffi- 
cient along a spanwise line from root 
to tip with steady data acquired after 
the unsteady transient had disap- 
peared. Note that the suction peak 
associated with the leading-edge 
vortex has moved outboard in the 
steady condition. 

The second figure presents 
photogrammetry data acquired 
during a run at Mach = 0.85, where 
the tip of the model was unre- 
strained as the pressure in the tunnel 
was slowly increased. Part (a) of the 
figure shows the time history of the 
deflection of the wingtip at three 
dynamic pressures just before the 
model became unstable; part (b) 
shows the phases and amplitudes of 
points on the wing compared with 
the trailing edge of the wingtip. The 
onset of instability was marked by a 
large increase in the amplitude of 
the motion, as well as an increase in 
the phase difference between the 
leading and trailing edges. 

This work is a first step in using 
PSP and photogrammetry to docu- 
ment complex, unsteady aeroelastic 
responses. Once these techniques 
are further refined, they can be used 
to better understand these responses 
and to help validate higher-order 
aeroelasticity methods. A natural 



Fig. 1. PSP image of flexible wing and corresponding pressure coefficient 
along x/C=0J5. 



46 



(i 1. H A [. C I V I I A \ 1 A T ION / Affordability 



Afro v a i t i r s a ^ d Space T r a s s p o r r a t i o .v 

T E C H \ L r E S T E R P R ! S E 



.6 
.4 
.2 

-.2 F 
-.4 - 



Tip deflection 




' t f = 48.4 Hz; C= ^.0004 

_ 6 I"" ■''''''''■''''■ I I I I M I I I I I I 




f = 50.3H2;C = 

_ 5 ri 1 t I n ] 1 M I I I I I I I I I ri r i I I I 




u 



l< 



lllflll ,==, 



Phase difference (deg) 
q = 3.33 



q = 3.45 



q = 3.47 



2 Hz; C= ^.0014 

" I M M M I I I I I I I I I I I I Ml 

.40 



.10 .20 .30 

Time (sec) 



(b) 



Amplitude ratio 






Fig. 2. Deflection of wingtip trailing edge, and phase difference and amplitude ratio with respect to tip trailing edge. 



extension of this work is to apply 
PSP and photogrammetry to tests of 
helicopter rotor blades. 

Point of Contact: E. Schairer/L. Hand 
(650)604-6925/1646 
eschairer@maiLarc.nasa.gov 
lhand@mail.arc.nasa.gov 



Visualizing Wind-Tunnel 
Experimental Data 

Samuel P. Uselton, Glenn Deardorff, 
Leslie Keely, Yinsyi Hung, Arsi Vaziri 

A new software tool that dis- 
plays data collected during wind- 
tunnel tests and that supports 
interactive analysis of those data has 
been developed at Ames Research 
Center. The tool, called exVis, 
allows visualization and interaction 
methods frequently used in analyz- 



ing computational results to be 
applied to experimentally collected 
data. This product is a step toward 
developing a unified environment 
for analyzing data from multiple 
sources. The ability of researchers 
and engineers to display and interact 
with data from a variety of sources 
within a single visualization environ- 
ment will increase the understanding 
gained from the data and reduce the 
time required. 

The initial product, exVis 1,0, 
was developed for data collected 



G 1. B A 1. (] [ \ i I. A V I A T I U 



I Affordahility 



47 



using a pressure-sensitive paint (PSP) 
system. The exVis software package 
reads a Flexible Image Transport 
System (FITS) format file produced 
by the PSP data-acquisition system 
which contains pressure values, and 
displays these data as an image in a 
window CImageViewer" shown at 
left in the figure (see Color Plate 9 in 
the Appendix)). The image can be 
displayed using a gray scale or a 
pseudocolor mapping of data to 
screen intensity. 

Users can select tools that allow 
the selection of a ''slice'' of data from 
the image along a line segment. This 
slice is then graphed in a separate 
window C'GraphViewer/' at the right 
in the figure) that shows how the 
coefficient of pressure varies along 
the length of the slice. Several slices 
can be selected from the same image 
and graphed on the same axes, or 
different slices (or sets of slices) can 
be graphed in separate windows. 
Multiple ImageViewers can each 
display an image, with selected 
slices graphed together or separately. 
The ImageViewer and GraphViewer 
can each be queried interactively 
to show the data value at the 
cursor; the GraphViewer also 
displays the corresponding location 
in the ImageViewer, The user can 
change the sizes and positions of 
ImageViewers and GraphViewers. 



Previously selected slices or graphs 
can be deleted, and a graph can be 
printed via postscript. A user can 
save the current status and restore 
the program to that state at a later 
time. An on-line help feature is also 
included. 

Although exVis was developed 
specifically with PSP data acquisi- 
tion in mind, it is sufficiently robust 
to work with other scalar data that 
are provided in FITS format, and 
could be easily adapted for data 
provided in other two-dimensional 
array formats. ExVis 1 .0 was com- 
pleted in July 1996 and integrated 
with other Ames-developed 
software for use in a wind-tunnel 
test (September 1996). 

Point of Contact: S. Uselton 
(650) 604-3985 
uselton@nas.nasa.gov 



Surface Tension Effects 
on Skin Friction 
Measurements 

G. Zilliac, A. Celic 

The Fringe-Imaging skin friction 
technique (FISF), which was origi- 
nally developed by Monson and 
Mateer at Ames Research Center 
and recently extended to three- 
dimensional flows, is the most 
accurate skin-friction measurement 
technique currently available. The 
principle of the FISF is that the 
skin friction at a point on an aero- 
dynamic surface can be determined 
by measuring the time-rate-of- 
change of the thickness of an oil 
drop placed on the surface under the 
influence of the external air bound- 
ary layer. Lubrication theory is used 
to relate the variation in oil-patch 
thickness to shear stress. 




Fig. 7. Oil-drop profile evolution. 



48 



Glob a i. C i v i l A v i a t i o n / Affordahility 



A E R S A i T I C S A S D S F A C E T R A S S P R T A T ! S 
T E C H S t G Y E S T E R P R I S E 



.1 l- 










- 






Dir>ni C mr>iil-t 








- 




-- 


— Simplified model 


~ 






E 
E 


- 








.01 
.0( 




t 1 1 Mil 1 


1 1 1 1 1 III 1 III 


1 1 III 1 11 1 


[)1 


.01 


.1 


1 10 






T, sec 





Fig, 2. Oil -drop thickness at X = 3 mm. 



The uncertainty of FISF measure- 
ments is estimated to be as low as 
4%, yet little is known about the 
effects of surface tension and wall- 
adhesion forces on the measured 
results. 

A modified version of the free- 
surface Navier-Stokes solver RIPPLE, 
developed at Los Alamos National 
Laboratories, was used to compute 
the time development of an oil drop 
on a surface under a simulated air 
boundary layer. RIPPLE uses the 
Volume of Fluid (VOF) method to 
track the surface and the Continuum 
Surface Force (CSF) approach to 
model surface tension and wall- 
adhesion effects. 

The figures show the develop- 
ment of an oil drop over a period of 
about 4 seconds. The profile of the 
drop (first figure) rapidly changes 



from its initial circular-arc shape to a 
wedge-like shape. The close agree- 
ment between the oil-drop thickness 
computed using RiPPLE and the 
results obtained with a greatly 
simplified numerical model of oil- 
drop flow (solution of Squires oil- 
drop equation which does not 
include surface tension and wall- 
adhesion effects) proves that surface 
tension will have a negligible effect 
on FISF results (second figure). 

Point of Contact: G. Zilliac 
(650) 604-3904 
gzilliac@mail.arc.nasa.gov 



Fullerene Gear Design, 
Simulation, and 
Visualization 

Albert Globus 

Ames Research Center has 
designed and computationally 
investigated a series of atomically 
precise gears based on fullerene 
chemistry (see first figure). These 
gears are approximately 1 nanom- 
eter in diameter and can turn at up 
to 100 gigahertz according to 
computer simulations. Long-range 
goals include the design and valida- 
tion of programmable molecular 
machines that can replicate and 
build useful aerospace products. 
This work will require massive 
computer power as nanotechnology 
research moves from ''what if to 
''how." The current simulations help 
in understanding these computa- 
tional requirements. If successful, 
molecular nanotechnology could 
revolutionize aeronautics and space 
technology by producing materials 
with radically improved strength-to- 
weight ratios and active materials 
filled with atomically precise 
machines that react to the environ- 
ment and that can self-repair. 

Hypothetical nanotechnology 
materials based on diamond chemis- 
try have great potential, but at 
present no clear path can be seen for 
their synthesis. In contrast, fullerenes 
(closed-cage carbon molecules) are 
now routinely produced in the 



G L ii A I. (^ [ V M. A V I A T 1 N / AffordabiUty 



49 




Fig. L Hypothetical carbon nanotube gears. Each gear is approximately 
1 nanometer in diameter Simulations suggest this system can rotate at up 
to 100 gigahertz. 




laboratory. The carbon form called 
C-60 (a molecule with 60 carbons 
arranged in a soccer-ball-like 
configuration) is produced commer- 
cially, and carbon nanotubes have 
been produced in bulk, some over 
1 00 microns long. It is possible to 
add a v^ide variety of molecular 
fragments to C-60 and other 
fullerenes. In particular, benzyne 
will react with the C-60 under mild 
laboratory conditions to form 
appended structures similar to 
gear teeth. 

A series of videos has been 
made including gear synthesis steps, 
favorable operating conditions, 
various gear configurations, a rack- 
and-pinion system (see second 
figure), and tooth slip conditions, 
which are all based on molecular 
dynamics simulations run on parallel 
computers. Each gear consists of a 
carbon nanotube with several 
O-benzyne fragments added to form 
gear teeth. Molecular dynamics 
simulations suggest that when one 
gear is rotated, intermolecular forces 
cause the second gear to rotate at up 
to 1 00 gigahertz without slippage or 
fragmentation. 

Point of Contact: A. Globus 
(650) 604-4404 
globus@nas.nasa.gov 



Fig, 2. Hypothetical rack-and-pinion system built from functionalized 
fullerene nanotubes. 



50 



Global Civil Aviation / AffordabiHty 



A t: R \ A IT I C S A S D S PAC E T R A S S P RTAT I S 
T E C H S L 6 Y E S T E R P R I S E 




Civil Tiltrotor Noise 
Abatement Approaches 

William A, Decker, 
Rickey C. Simmons 

A civil tiltrotor regional transport 
operating from vertiports near urban 
and business centers offers consider- 
able potential for relieving conven- 
tional airport runway congestion. 
Flight operations near urban areas 
must also minimize noise effects 
vt'hile providing for all-weather 
schedule reliability. Conversion from 
airplane-mode flight to a helicopter- 
like landing while minimizing noise 
presents both a challenge and an 
opportunity for tiltrotor aircraft. The 
introduction of worldwide aircraft 
position information with the Global 
Positioning Satellite system provides 
an economic guidance source to 
support the desired low-noise, 
all-weather operations. 

Initial tiltrotor noise-abatement 
landing approach profiles were 
constructed based on actual noise 
measurements of the XV-1 5 tiltrotor 
research aircraft. Instead of a 
conventional landing approach 
which intercepts and tracks down a 
3-degree glideslope, two alternative 
profiles were constructed, both 
featuring an initial 3-degree descent 
which changes to a steep 9-degree 
descent close to the landing spot 
(first figure). Based on the XV-1 5 
noise data, the steeper final 
approach path offers good potential 
for reducing noise while providing 
better clearance of obstacles near 



Nacelle deceleration 
for landing from 
Landing Decision Point 
at 200 ft altitude 



1 10/60 1^40/30 180/0 <- 250/0 

2000 ft altitude 



250/0 




.t^ 



3000 ft altitude 



Fig. 1. Tiltrotor noise-abatement approach profiles with two-segment 
flightpaths. Altitude (flightpath angle), airspeed, and nacelle angle (for 
example, 110/60'') are specified as functions of distance from landing. 



the vertiport. Depending on the 
terrain and airspace constraints 
associated with nearby conventional 
airports, the initial approach seg- 
ment involved conducting either a 
level-flight conversion toward the 
helicopter-mode landing configura- 
tion, or performing all of the conver- 
sion in descending flight. Both 
approach profiles carefully specified 
airspeed, descent rate (flightpath 
position) and proprotor tilt angle 
based on the distance from the 
landing spot. 

Cockpit displays and control 
features have been developed to 
help guide pilots to reliably and 
easily fly these complex approach 
profiles in all-weather conditions. As 
illustrated in the second figure, the 
displays were adapted from modern 



electronic display formats for 
transport aircraft. Pursuit guidance 
using flightpath vector symbols, 
adapted from military head-up 
display designs, has been developed 
for tiltrotor use and applied to the 
panel-mounted electronic displays. 

The unique large-motion cueing 
capability of the Vertical Motion 
Simulator (VMS) at Ames Research 
Center contributed to piloted 
evaluations of approach profiles 
and displays. Use of a visual scene 
based on the San Francisco Bay 
Area, developed in the VMS 
SIMLAB, contributed to the evalua- 
tion scenario and provided geo- 
graphic rationale for the complex 
approach flightpath. Eleven pilots 
representing NASA, the FAA, British 
CAA, and the rotorcraft industry 



G [, 



** '^ '^ '- G i V i !, A V i A T ION / Hnr iron mental Co^npntihil ity 



51 




Fig. 2. Pilot's view on approach to urban vertiport. Electronic cockpit 
displays provide guidance for complex noise-abatement approach profiles. 



evaluated the approaches and 
displays, and provided comments 
for further refinement. 

Point of Contact: W. Decker 
(650) 604-5362 
bdecker@mail.arc.nasa.gov 



sons with similar wind-tunnel 
measurements, (2) to use these 
comparisons to better understand 
BVI noise, and (3) to determine how 
to minimize BVI noise on rotorcraft. 
During the flight portion of this 
test, the YO-3A fixed-wing aircraft 
used a wing-tip-mounted micro- 



phone to measure BVI noise gener- 
ated by the XV-1 5 tiltrotor. Four 
flights were performed with the 
YO-3A and XV-1 5 flying in close 
formation, as shown in the figure. 
The flight conditions and micro- 
phone location were chosen to 
measure the most prominent BV! 
noise. The starboard wing-tip 
microphone of the YO-3A was 
positioned 20 degrees below the 
right rotor hub of the XV-1 5 and 
offset to the right to pick up BVIs 
occurring at a rotor azimuth of 
1 50 degrees (see figure). The dis- 
tance between the starboard wing- 
tip microphone and the rotor hub 
was 3 rotor diameters {7S feet). 
Target flight conditions included 
advance ratios of 0.1 65 and 0.1 85, 
tip Mach number of 0.69, a rotor 
weight coefficient of 0.01 1 1 , and 
descent rates of 300-1 1 00 feet per 
minute. 

The results of the flight test were 
used to determine equivalent test 
conditions for a series of tests of the 
XV-1 5 rotor in the National Full- 
Scale Aerodynamics Complex 



XV-1 5 Blade-Vortex 
Interaction Noise 

C. W. Acree, Megan S, McCluer, 
Cahit Kitaplioglu 

As part of the continuing 
In-Flight Rotorcraft Acoustics 
Program, flight tests of the XV-1 5 
tiltrotor were completed and wind- 
tunnel tests are under way. Program 
objectives are (1) to use the NASA 
Ames YO-3A acoustic research 
aircraft to measure blade-vortex 
interaction (BVI) noise from rotor- 
craft in flight and to make compari- 




Fig. /. YO-3A and XV'15 in formation flight. 



52 



G !. H A I 



C 1 \' 1 [. A \' 1 A T 1 



N / Enviroumental Compatihiliiy 



A E R S A I r ! C S i S D S P A C E T R A S S F R T A T I S 
T E C H S L C y E \ T E R P R ! S E 



(NFAC) 80- by 1 20-Foot Wind 
Tunnel test section at Ames Research 
Center. A single (right-side) rotor was 
mounted on the Rotor Test Appara- 
tus (RTA), and microphones were 
placed in locations geometrically 
identical to the relative positions of 
the YO-3A microphones and XV-1 5 
rotor during flight. The data from the 
NFAC tests have been compared 
with the flight data to validate the 
wind-tunnel test and the measure- 
ment methods. Although very good 
qualitative agreement was achieved, 
good quantitative agreement has yet 
to be demonstrated. In particular, the 
ability to match wind tunnel with 
flight will likely require very good 
characterization of the flight operat- 
ing conditions to enable matching of 
the exact test condition in the wind 
tunnel. 

Point of Contact: C. Acree 
(650) 604-5423 
wacree@mail.arc.nasa.gov 



Tiltrotor Aeroacoustic 
Model 

Larry Young 

NASA conducts tiltrotor 
research programs in order to meet 
national requirements for military 
and civilian tiltrotor aircraft. This 
research requires moderate-to-large- 
scale wind-tunnel testing of tiltrotor 
models. Such testing provides the 
data necessary to confirm aero- 
acoustic prediction methods, and 
to investigate and demonstrate 
advanced civil tiltrotor and high- 
speed rotorcraft technologies. 

In 1991, NASA established that 
it had a requirement to gain an 



improved understanding of the 
aeroacoustic characteristics of 
tiltrotor aircraft and, therefore, 
initiated the development of two 
hardware-compatible test rigs: an 
isolated-rotor test stand and a full- 
span model (dual rotors with a 
complete airframe representation). 
These two test stands are inclusively 
called the Tilt Rotor Aeroacoustic 
Model (TRAM). The isolated rotor 
test stand is not a stand-alone unit, 
but is instead intended to be com- 
posed chiefly of major subassem- 
blies of the full-span configuration. 

The TRAM will be used as a test 
bed for testing moderate-scale tilt- 
rotor models in two different test 
configurations in different research 
facilities: (1 ) isolated rotor testing at 
the Duits-Netherlands Windkannal 
(DNW) in The Netherlands; and 
(2) isolated rotor and full-span 
testing at the National Full-Scale 
Aerodynamics Complex (NFAC) at 
Ames Research Center. 

A checkout test of the TRAM 
isolated-rotor test stand was con- 



ducted in the NFAC N246 Model 
Preparation building at Ames. 
The checkout testing included the 
acquisition of hover data — including 
rotor airloads data — for a 1/4-scale 
set of proprotor blades using the 
NFAC NPRIME data-acquisition 
system. Key test-stand capabilities 
being checked out include a rotating 
amplifier system (RAS) developed 
by the Nationaal Lucht-en 
Ruimtevaartlaboratorium (NLR) 
and the Monitoring Front-End Data 
System (MFEDS) real-time safety- 
of-flight monitoring system. A 
considerable amount of infrastruc- 
ture buildup — including armor-plate 
shielding and control room installa- 
tion — was required in the N246 
building to enable TRAM functional 
testing. 

The full-span version of the 
TRAM test stand is also being 
concurrently developed. Fabrication 
of full-span model development is 
under way (see figure). A new 
generation rotor-control console for 




Fig, 1. Full-span TRAM fabrication and assembly. 



(i 1. {> n A L (] i V M. A V i A T ION / R H r I r luu c u t (i I Compatihilily 



53 



the isolated rotor test and the full- 
span model is also in development. 
The 1/4-scale TRAM rotors and 
airframe are based on the design of 
the V-22 Osprey tiltrotor aircraft. 
The TRAM test stand will also be an 
advanced technology demonstrator 
platform for the Short Haul Civil 
Tiltrotor (SH(CT)) program, a sub- 
element of the Advanced Subsonic 
Technology (AST) initiative. Both 
Boeing and Sikorsky Aircraft have 
been contracted by the NASA 
SH(CT) program to develop interface 
hardware to test a new generation of 
small-scale, advanced proprotors on 
theTRAM test stand. 

Point of Contact: L. Young 
(650) 604-4022 
lyoung@maiLarc.nasa.gov 



Predicting and Analyzing 
Rotorcraft Noise 

Roger C. Strawn^ Rupak Biswas^ 
Lenny Oliker 

Aerodynamic noise is a major 
problem for future civil tiltrotor 
designs. Noise is a particular prob- 
lem during landing when the rotor- 
wake system is pushed up into the 
rotating blades. The resulting blade- 
vortex interaction (BVI) noise can 
be unacceptable in urban environ- 
ments. All four of the major U.S. 
helicopter companies have active 
research programs with the objective 
of reducing tiltrotor and helicopter 
BVI noise. 

Three things are required in 
order for rotor designers to system- 
atically reduce BVI noise: (1 ) an 
improvement in the accuracy of far- 
field noise predictions; (2) the ability 



to analyze those predictions in a 
way that identifies the sources of the 
offending noise; and (3) control of 
rotor-blade motion in ways that 
reduce the noise. Researchers at 
Ames Research Center have recently 
developed new acoustic prediction 
and analysis tools that directly 
address the first two of these three 
objectives. 

The new acoustic prediction 
scheme combines computational 
fluid dynamics (CFD) with Kirchhoff 
integration methods. The CFD 
methods compute the nonlinear 
aerodynamics and acoustics close to 
the rotor blades, and the Kirchhoff 
integrations carry the acoustic 
information to the far-field in a 
computationally efficient manner. 
These CFD/Kirchhoff methods run 
on the IBM SP-2 parallel computer 
to facilitate the computation of time- 
accurate acoustic information over 
large regions of space. Computer 
audio and visual rendering (see 
figure) of the resulting far-field 



acoustic propagation allows rotor 
designers to "see and hear" the noise 
in the far-field. Computer animations 
of the far-field noise propagation 
convey much more information than 
can be obtained from typical rotor- 
noise experiments, in which the 
data are limited to those picked 
up by a few far-field microphones. 
In addition to the new prediction 
and visualization tools, the CFD/ 
Kirchhoff analysis can be modified 
to highlight the sources of noise on 
the Kirchhoff surface. This allows a 
rotor designer to pinpoint the origins 
of the far-field noise and to modify 
the blade trajectories to minimize 
the blade-vortex interactions. 

The new prediction and analysis 
tools will have applications to a 
variety of helicopter and tiltrotor 
geometries. 

Point of Contact: R. Strawn 

(650)604-4510 

rstrawn@maiLarc.nasa.gov 




Fig. 1. Computed acoustic pressure contours for the AH-1 Cobra helicopter. 



54 



C; 1. {) H A I, C 1 V I L A V 1 A r I (J N / Enrironmtnilal CompntihUily 



A f. R \ A i' T I c s A \ D Space T r a \ s p o r t a t i o s 

T f C H S L a Y EsTERPRiSE 



Airframe Noise 
Measurements: 
Atmospheric Pressure 

W. Clifton Home, Julie A. Hayes, 
Michael E. Watts, Paul H. Bent 

With the trend toward quieter, 
higher bypass-ratio engines in 
commercial transport aircraft, 
airframe noise is a significant 
problem during approach and 
landing. In support of the Noise 
Reduction Element of the Advanced 
Subsonic Technology (AST) 
Program, an aeroacoustic test of an 
unpowered, 4,7%-scale McDonnell 
Douglas DC-10 was conducted in 
the NASA Ames 40- by 80- Foot 
Wind Tunnel (see figure). Acoustical 
measurements included streamwise 
traverses of four microphones, far- 
field measurements with directional 
microphones using parabolic 
reflectors, and noise-source location 
with a 40-element phased micro- 
phone array. Aerodynamic measure- 
ments included steady and unsteady 
wing-surface pressures and model 
incidence. 

The primary objectives of this 
test were to (1 ) assess the feasibility 
of extrapolating wind-tunnel air- 
frame noise measurements for 
comparison with full-scale flight-test 
results, (2) locate prominent noise 
sources with the acoustic array, 
and (3) evaluate prototype noise- 
reduction concepts on a realistic 
transport configuration. 

Measurements supporting all 
of the primary test objectives were 
obtained. In particular, acoustic 
images of airframe noise sources on 
the high-lift system and landing gear 
were obtained. This was achieved 
using the Microphone Array Phased 




Fig. 1. DC-W model in the Ames 40- by 80-Foot Wind Tunnel. 



Processing System developed at 
Ames Research Center. 
This system used high-performance 
graphical workstations in the control 
room to process the array data and 
display results during the test. The 
system processed 200 frequencies at 
each data point within 30 minutes, 
thus allowing all test data to be 
processed during the test (as much 



as 3 months of posttest processing 
were required in previous tests). The 
processing system was designed to 
fully support existing remote access 
wind-tunnel capabilities. 

Studies of flap-edge fences 
designed to reduce noise emanating 
from the flap edge showed broad- 
band noise reduction levels of 



(i r 1^ A I, C I V I L A V I A T [ N / Eh V I v u m e if t a I Compalibilily 



55 



3-4 decibels. Correlation of the 
noise signals on a far-field micro- 
phone with the unsteady pressure 
signals on the nnodel aircraft surface 
was effective in locating the source 
of the airframe noise. 

Point of Contact: C. Home 
(650) 604-4571 
chorne@mail.arc.nasa.gov 



Airframe Noise 
Measurements: Pressures 
to 4.7 Atmospheres 

W. Clifton Home, Stephen M. Jaeger, 
Mahendra Joshi, James R. Underbrink 

Airframe noise, prinnarily 
attributable to the landing gear, 
leading-edge slat, and trailing-edge 
flap, is known to be a significant 
component of the approach and 
landing noise of commercial trans- 
port aircraft. The NASA Advanced 
Subsonic Technology (AST) Noise 
Reduction Technology Program is 
sponsoring several measurement and 
prediction programs in order to 
reduce airframe noise by 4 decibels 
below 1992 levels. Because of the 
expense of flight tests, an important 
element of this research is the 
application of new measurement 
technology, such as the multisensor 
phased acoustic array, on tests of 
small-scale wind-tunnel models and 
then scaling and extrapolating these 
measurements to full-scale. The 
physics of aerodynamic noise 
generation and the effect of geomet- 
ric scale are highly complex and not 



well understood. Therefore, acoustic 
measurements, over a range of scale 
factors (or Reynolds numbers), are 
required to identify the important 
noise sources on full-scale aircraft 
and to determine the scaling laws 
that will permit accurate full-scale 
predictions from small-scale tests. 

Recent successful tests by Ames 
and Boeing using a multielement 
phased microphone array in closed, 
untreated test sections suggested that 
useful airframe noise measurements 
could be obtained in the Ames 
1 2-Foot Pressure Wind Tunnel 
(PWT) for variable Reynolds number 
testing. Feasibility tests of a Boeing 
52-element array in the 12-Foot 
PWT were planned in two phases. 
An initial study of the operational 
feasibility was planned with the test 
section empty except for a survey 
strut, model ground plane, and 
calibration noise sources. If this 
initial study proved to be successful, 
a second demonstration of aero- 
acoustic measurements of a 
4.7%-scale DC-10 was planned in 
conjunction with McDonnell 
Douglas to acquire the first variable 
Reynolds number airframe noise 
measurements. The results would be 
compared with atmospheric mea- 
surements acquired from the same 
model in the 40- by 80-Foot Wind 
Tunnel at Ames with a 40-element 
phased-array designed and 
assembled at Ames. 

The first feasibility test demon- 
strated that the microphone-array 
system (sensors, cabling, signal 
acquisition and processing, and 
computer networking) could be 



accommodated with existing facility 
support infrastructure. Several 
aeroacoustic noise sources were 
located with the array, associated 
with support struts and the trailing 
edge of the semi-span ground 
plane. The second test utilizing the 
4.7%-scale DC-10 aircraft model 
was then undertaken. During this 
second test, the Boeing array 
acquired noise measurements to 
60 kilohertz, corresponding to about 
3 kilohertz full-scale, and identified 
noise sources at atmospheric 
pressure which corresponded 
closely with sources identified in 
the previous 40- by 80-Foot Wind 
Tunnel test (see figure (see Color 
Plate 1 in the Appendix)). In 
addition, noise measurements 
were obtained for pressures up to 
4.7 atmospheres. Preliminary 
evaluation of these measurements 
show that the relative strength of the 
noise sources varies with Reynolds 
number in a complex and as yet 
unpredictable manner, stressing 
the need for large-scale, high- 
Reynolds-number validation of 
noise-prediction methods. 

Point of Contact: C. Home 
(650) 604-4571 
chorne@maiI,arc.nasa.gov 



56 



G I B A I C I V I L A V ! A T I K / Environmental Compatihiliiy 



A E R S A i T I C S A S D S P A C E JrAHSFORTATIOS 

Techsolocy E S T E R P X I s e 




X-36 Pioneers Advanced 
Aerodynamics and Flight 
Controls 

Rodney Bailey, Mark Sumich 

The X-36 project is pioneering 
the development of advanced 
aerodynamics and flight control 
technologies that will permit the 
concurrence of low radar observa- 
bility, agility, and supersonic speed 
in a single design. Prior to the X-36 
technologies, these features were 
mutually exclusive. The flight 
evaluations of the X-36 aircraft, a 
joint NASA/McDonnell Douglas 
project, are currently under way at 
the Dryden Flight Research Center. 

The project began in February 
1 994 after several years of research 
and concept development in the 
Ames facilities by both Ames and 
McDonnell Douglas engineers. 



The project involves the design, 
fabrication, and flight evaluation of 
two unmanned aircraft. The 10-foot- 
span flight articles are about 1 8 feet 
long and 3 feet high and weigh 
1250 pounds. 

Features of the X-36 include 
multifunction wing trailing edge 
control surfaces, combined ailerons, 
split ailerons and flaps, wing leading 
edge flaps, and all moving canards. 
The aircraft has neither vertical nor 
horizontal aft tail surfaces. The 
aircraft (shown in the figure) is 
naturally unstable in both the pitch 
and yaw directions, with artificial 
stability provided by the control laws 
utilizing both aerodynamic and 
propulsion forces. 

The flight testing at Dryden will 
evaluate real-time stability margin 
and will determine aerodynamic, 
stability, and control data on the 
configuration using the parameter- 
identification derivative. 

Point of Contact: R. Bailey 
(650) 604-6265 
rbailey@mail.arc.nasa.gov 




Closed-Loop Neural 
Control of Rotorcraft 
Vibration 

Sesi Kottapalli 

A study on identification and 
control of rotorcraft hub loads 
(vibration) using neural networks 
has been initiated at Ames Research 
Center. Neural control simulation 
covers both identification and 
control of hub loads, where a metric 
is used to represent the hub load 
components. The closed-loop 
controller must be computationally 
efficient, it must converge quickly 
(in six iterations or less), and 
gradient-based methods must 
not be used. 

The program objective is to 
implement closed-loop neural 
control of rotorcraft vibration in 
a wind tunnel in two phases: 
(1) conducting neural-control 
simulations using existing individual 
blade-control wind-tunnel test data, 
and (2) implementing closed-loop 
neural control during a future full- 
scale rotorcraft wind-tunnel test 
(contingent on first-phase results 
being sufficiently promising). 

The identification (plant model- 
ing) procedure, which involves the 
use of a two-hidden-layer radial 
basis function neural network, has 
been completed- The procedure has 
been extended to multi-input, multi- 
output applications. A simple, 
straightforward-to-implement neural- 
control technique, the "direct 
inverse'' method, was selected, 
successfully applied, and found to 



Fig. 1 . X-36 Tailless Fighter Agility Research Aircraft prepares for flight at the 
Dryden Flight Research Center. 



V 1, [ \ i 



N A R Y T !•: c H N 1, (; \ ]. }■; A PS / hniorative Tccbnoloi^y and Tools 



57 



Present Closed Loop Neural Control Simulation 
("metric" refers to vibratory hub loads metric) 



Control 
phase Input 



Actual 

metric 



Desired 
metric 




Half-Interval 

method 

(user specified 

metric reduction) 



Updated 
control phase 
input 



Inverted 
neural network 

for control 
(supplies new 
control Input) 



Fig. 1 . Inverted neural network for vibration control. 



be robust. The present approach (see 
the figure) has the following essential 
ingredients: accurate plant model- 
ing, halving of the metric in order to 
accelerate controller convergence, 
'Inverted-axes" control modeling, 
and a feedback iterative loop. A 
back-propagation neural netv^ork 
used in the control step has been 
successful. Current results shov^ that 
a 66% reduction in baseline rotor 
vibratory hub loads is achievable in 
only four controller iterations. A 
limited-scope comparison of the 
results from the present neural- 
control procedure v^ith those from 
a one-step deterministic controller 
showed that the two control meth- 
ods are comparable, with neural 
control being more robust. 

Point of Contact: S, Kottapalli 
(650) 604-3092 
skottapalli@mail.arc.nasa.gov 



Coupled Navier-Stokes 
and Optimizer Analysis 
of a Transonic Wing 

Roxana M. Greenman, 
Samson Cheung, Eugene L. Tu 

Numerical optimization can be 
used as a means of studying the 
effects of the geometric parameters 
on the aerodynamics of a wing. 
Different types of aerodynamic 
numerical-optimization techniques 
that have been researched in recent 
years have been found to need 
improvement. New optimization 
tools must be developed to provide 
efficient means for improving the 
design process and understanding 
the flow physics. The overall objec- 
tive is to develop the necessary tools 
needed to help, using computational 
fluid dynamics for the design of 
advanced wings through the inclu- 
sion of new, promising optimization 



modules to the code OVERFLOW. 
The object is to demonstrate the use 
of a numerical-optimization routine 
to understand the significance of 
different geometric parameters on 
the aerodynamics of a wing. 

This study demonstrated how 
one optimizer package is used to 
study the flow physics of a transonic 
wing. This numerical-optimization 
routine is integrated with a grid 
generator and a validated Navier- 
Stokes flow solver. The effects of 
wing sweep angle, root and tip twist 
angles, dihedral, and angle of attack 
on the lift-to-drag ratio are studied 
using surface-pressure contours and 
pressure distributions. The results 
show that when lift-to-drag ratio 
is increased using single design 
variable optimization, the resulting 
lift coefficient falls below the 
original value. The angle of attack 
must be increased in order to 
preserve the original lift coefficient 
with the modified geometry. A better 
way to find the optimal configura- 
tion with the highest lift-to-drag ratio 
is by first optimizing with multiple 
design variables (see the figure (see 
Color Plate 1 1 in the Appendix)), 
then including the angle of attack in 
the design loop with lift constrained 
to the original value. This method 
showed an improvement of 2.4% 
(with constant lift) in the lift-to-drag 
ratio. 

Point of Contact: R. Greenman 
(650) 604-3997 
rgreenman@mail.arc.nasa.gov 



58 



R K V 1. [1 T I M A H Y T K c H N 1. i\ V L K A i* s / Innovutive Technology and Tools 



A E R S A L! n C S A S D S P A C E T R A S S F R T A T ! S 
T E C H S L G Y E S T E X F R ! S E 



Parallel Unstructured 
Mesh Adaption 

Rupak Biswas, Leonid Oliker, 
Roger Strawn 

Computational methods to solve 
large-scale realistic problems can 
be made more efficient and cost 
effective by using them in conjunc- 
tion with dynamic mesh adaption 
procedures that perform coarsening 
and refinement to capture interesting 
solution features. Such adaptive 
procedures evolve with the solution, 
and they provide scientists a robust 
and reliable methodology to obtain 
solutions on adapted meshes that are 
comparable to those obtained on 
globally fine grids but at a much 
lower computational cost. In addi- 
tion, with the popularity of distrib- 
uted computing, parallel 
unstructured mesh adaption proce- 
dures are absolutely imperative. 

An efficient solution-adaptive 
procedure has been developed for 
the simultaneous coarsening and 
refinement of unstructured tetrahe- 
dral meshes. An innovative data 
structure, which uses a combination 
of dynamically allocated arrays and 
linked lists, allows the mesh connec- 
tivity to be rapidly reconstructed 
after individual points are added 
and/or deleted. The data structure, 
based on edges of the mesh rather 
than the tetrahedral elements, not 
only enhances the efficiency but 
also facilitates anisotropic mesh 
adaption. This scenario means that 
each tetrahedral element is defined 
by its six edges rather than by its four 
vertices. Each edge, in turn, main- 
tains a list of all the elements that 
share it. Thus, elements that need to 



be modified can be rapidly identified 
when an edge is inserted or deleted. 
Regions of the mesh that need to be 
adapted are identified by marking 
the edges based on an error indica- 
tor. The adaption procedure for each 
element is determined from a binary 
pattern that depends on which of its 
six edges are marked. The edge data 
structures and pattern marking 
ensure that the refinement and 
coarsening procedures can be 
performed efficiently. 

The parallel mesh adaption 
procedure has been used to model 
the high-speed impulsive noise for a 
UH-1 H helicopter rotor in hover. A 
coarse mesh is initially generated 
and additional points are dynami- 
cally added to capture the acoustic 
wave and the shock near the tip. 
Mesh coarsening is used to remove 
unnecessary points from regions 
with smooth solutions; thus, existing 
grid points are redistributed in a 
more efficient manner. A close-up of 
the final grid and pressure contours 
in the plane of the rotor after three 
adaption steps is shown in the figure 
(see Color Plate 1 2 in the Appendix). 
Results show excellent agreement 
with experimental microphone data 
for far-field acoustic pressure. 
Parallel speedup depends on the 
fraction of the mesh that is adapted. 
When 33% of the mesh was refined, 
a speedup of 43. 2X was obtained on 
64 processors. The speedup im- 
proved to 51 .5X when 60% of the 
mesh was refined. 

Current results demonstrate 
significant gains in computational 
efficiency when the parallel 
solution-adaptive method is com- 
pared to its fixed-grid counterparts. 



These improvements are necessary 
for future high-fidelity simulations of 
realistic problems in unsteady 
environments. 

Point of Contact: R. Biswas 

(650)604-4411 

rbiswas@nas.nasa.gov 



Load Balancing Adaptive 
Unstructured Meshes 

Rupak Biswas, Leonid Oliker, 
Andrew Sohn 

Mesh adaption is a powerful 
tool for efficient, unstructured-grid 
computations, but it causes load 
imbalance among processors on a 
parallel machine. A novel method 
has been developed to dynamically 
balance the processor v^orkloads 
with a global view. A dual graph 
representation of the initial computa- 
tional mesh keeps the complexity 
and connectivity constant during the 
course of an adaptive computation. 
It is a portable system for efficiently 
performing large-scale adaptive 
calculations in a parallel message- 
passing environment. 

The first figure depicts the 
framework used for parallel adaptive 
scientific computations. Mesh 
adaption, repartitioning, processor 
assignment, and data remapping are 
critical components of the frame- 
work that must be accomplished 
rapidly and efficiently so as not to 
cause a significant overhead to the 
numerical simulation. The unstruc- 
tured mesh is first partitioned and 
mapped among the available 
processors. A solver then runs for 
several iterations, updating solution 



R K V {) I (1 T [ N A R y T [-: c ij k o i. o o y L k a p s / 



Innovative Technology and Tools 



59 



Initialization 


Initial mesh 


1 


r 


Partitioning 


1 


r 


Mapping 






1 





Mesh adaptor 



Edge marlcing 



Coarsening 



Solver 



Refinement 




Repartitioning 

} 



Reassignment 

T 

— Expensive? 



Remapping 

T 



O' 



Fig. 1. Framework for parallel adaptive scientific computation with load 
balancing. 



10' 



10' 




?'° Po-^ 



10" 



.2 -2 

110 2 

o 

s 

10 ^ 



10" 



10 



.-5 



-L 



16 



24 32 40 

Number of processors 



-O Adaption 

-CD- Partitioning 

-V- Reassignment 

-A- Remapping 

I I I 

64 



48 



56 



Fig, 2. Anatomy of total execution time. 



variables. After an acceptable 
solution is obtained^ the mesh 
adaption procedure is invoked. It 
first targets edges of the mesh for 
coarsening and refinement based on 
an error indicator computed from 
the numerical solution. The old 
mesh is then coarsened, resulting in 
a smaller grid. Since edges have 
already been marked for refinement, 
it is possible to exactly predict the 
new mesh before actually perform- 
ing the refinement step. Program 
control is thus passed to the load 
balancer. If a quick evaluation step 
determines that the new adapted 
mesh will be unacceptably load 
balanced, a repartitioner divides the 
mesh into subgrids. The new parti- 
tions are then assigned to the 
processors in a way that minimizes 
the cost of data movement. If the 
remapping cost is less than the 
computational gain that would be 
achieved with balanced partitions, 
all necessary data are appropriately 
redistributed; otherwise, the new 
partitioning is discarded. The 
computational mesh is then actually 
refined and the numerical calcula- 
tion is restarted. 

An extensive study with a 
computational mesh used to simu- 
late a UH-1 H helicopter rotor blade 
acoustics experiment indicated that 
the load-balancing scheme is 
effective for large-scale scientific 
computations on distributed-memory 
machines and scales excellently with 
the number of processors. The 
second figure shows how the total 
execution time was spent in each 
module of the framework when the 
mesh was refined to be more than 
five times larger than the original. 



60 



R F V {) L U T I N A H Y T I- C H N 1. G 



V L F, A PS / Innovative Technology and Tools 



A E K \ A I T I C S A S D S P A C E T R A S S P R T A T I S 
T E C H S L C Y E S T E R P R I S E 



The processor reassignment times 
are almost negligible compared to 
the other times. The repartltioning 
time is generally independent of the 
number of processors. Both the 
adaption and remapping times 
gradually decrease as the number 
of processors is increased. These 
sample results show that none of 
the individual modules will be a 
bottleneck on massively parallel 
machines. 

Point of Contact: R. Biswas 

(650)604-4411 

rbiswas@nas.na$a.gov 



Initial Release of the Field 
Encapsulation Library 

Steve Bryson 

Computational fluid dynamics 
(CFD) simulations are a new way of 
designing and evaluating aircraft 
designs on a computer, CFD simula- 
tions are computed at many points, 
forming a three-dimensional grid in 
space. The variety of grid types can 
greatly complicate the task of using 
computer graphics visualization to 
investigate the CFD simulations (see 
the figure (see Color Plate 1 3 In the 
Appendix)). Visualization algorithms 
can depend very strongly on the type 
of grid; an algorithm written for one 
grid type may not work on another 
grid type. This problem is particu- 
larly acute when developing a 
general-purpose visualization system 
for CFD. The objective of the Field 
Encapsulation Library (FEL) project 
is to solve the grid-dependence 



problem by implementing a library 
that provides a grid-independent 
approach to CFD data access. Using 
FEL, a single visualization algorithm 
can work on a variety of grid types. 
Very high performance data access 
is an additional objective in order to 
support interactive visualization, 
which allows intuitive exploration 
of the simulated flow phenomena. 
Another objective of FEL is to 
provide programmers with the ability 
to access, organize, and manipulate 
field-type data on many types of 
numerical grids at the same time. 

FEL was fully implemented and 
released for structured multizone 
and unstructured grid types, proving 
the grid-independence concept. 
Data access times averaging 
0.0002 second per access were 
measured. These capabilities 
allow the development of high- 
performance visualization algorithms 
that will work for all supported grid 
types. FEL has been integrated into 
the Virtual Windtunnel as well as 
other individual visualization 
systems. The concepts behind 
FEL were presented at Visualization 
'96, an IEEE peer-reviewed confer- 
ence held in October 1996 in 
San Francisco, California. 

FEL is an example of a ''horizon- 
tal product/' a set of software 
libraries that can be used in a variety 
of applications, significantly reduc- 
ing the time required to implement a 
new visualization system. 

Point of Contact: S. Bryson 
(650) 604-4524 
bryson@nas.nasa.gov 



RANS-MP: Portable 
Parallel Navier-Stokes 
Solver 

R. Van der Wijngaart, 
Maurice Yarrow 

Efficient, cost-effective solution 
of flow problems of interest to the 
aircraft industry requires competi- 
tively priced hardware and maintain- 
able, portable software. The current 
generation of parallel computers 
provides that hardware, and stan- 
dard system software such as the 
message passing interface (MPI) 
library provides half of the software 
solution. The other half comes from 
application programs that employ 
algorithms that are tolerant of 
differences in specific machine 
parameters, most notably the speed 
with which the processors in a 
parallel computer can exchange 
data. The Reynolds-averaged 
Navier-Stokes with multipartitioning 
(RANS-MP) program has been 
specifically designed to meet these 
requirements. 

The newly developed RANS-MP 
flow solver program employs the 
advanced, bidirectional, multi- 
partition algorithm. This method 
limits data movements between 
processors to a minimum, rendering 
it insensitive to the peculiarities 
of different parallel computers. 
RANS-MP uses the widely available 
MPI, which means it can run 
efficiently on most of today's parallel 
machines. It also employs a pilot 



R K 



^ *' ■• ^' '■ ' f> >^ -^ ^ V T !■: c II N L (; v L h a p s / hniovative Technology and Tools 



61 



version of MPI's companion input/ 
output (I/O) library MPI-I/O, which 
allows efficient scheduling of 
previously prohibitively expensive 
parallel read and write operations. 

The first figure shows the 
performance of the program on the 
Ames Research Center IBM SP-2 
parallel computer in terms of 
millions of points updated per 
second by the flow solver. The 
problem is that of viscous flow 
around an aircraft wing. A good 
measure of the quality of a parallel 
program is its scalability, which is 
defined as the improvement in 
performance as more processors 
are added to the computation. The 
figure indicates that, even for a 
relatively small grid of fixed size of 
200,000 points, the scalability is 
quite good (heavy dashed curve). 
Performance is even better if the 
problem size is allowed to grow with 
the number of processors employed 
(heavy solid curve). For reference, 
the ideal speedup curve (thin solid 
line) is also shown. 

The second figure indicates the 
improvement in performance of 
using the MPI-l/O library over that of 
the standard, nonparallel I/O library 
on the SP-2, where a gain of up to a 
factor of 30 in write speed can be 
observed. 



Although RANS-MP primarily 
targets cache-based systems that 
make up the bulk of the latest high- 
performance computing architec- 
tures, care has also been taken to 
provide good performance on more 
traditional computers such as the 
Cray C90 and the NEC SX-4. There- 



fore, the code must be written to 
allow efficient vectorization, the 
main source of performance 
improvement on such machines. 
When the wing problem is solved on 
the C90, RANS-MP attains a respect- 
able 45% of the maximum possible 
performance of the machine. 



Variable grid size 




Fixed grid size 



I . , , I . ■ ■ ■ 1 I I I I i I I I I I I I I M 1 



20 40 60 80 100 120 

Number of processors 



Fig. /. Millions of points updated per second on IBM SP-2. 



62 



R K V I V T I N A K Y T K C H N 1. (i V L F A i' S 



/ Innovative Technology and Tools 



A E R \ A V T I C S A S D S P A C E T R A \ S P R T A T I S 
T E C H S i 6 Y E S T E R P R I S E 



2.00 














/ ^ 


\ 






1.00 
cr> 

1 

(0 

1. 


^ 


V 


\___ MPI-I/O 




..___ 






1 










.10 


- 










.05 


1 


1 1 1 ( 1 1 


^ - - -. ^ ^ Standard I/O 

1 1 1 1 1 I 1 1 1 ) 1 


1 1 r 1 1 


~i~ r T t 







20 


40 60 


80 


100 






Number of processors 







Fig. 2. Write speed in megabytes per second for MPI-I/O and standard I/O 
on IBM SP-2, 



Measurements show that more than 
99% of all computational statements 
are executed in vector mode. 

Point of Contact: M. Yarrow 
(650) 604-5708 
yarrow@nas.nasa,gov 



Parallel Tools for Parallel 
and Distributed Computer 
Systems 

D. DiNucci, M. Frumkin, R. Hood, 
H. jin, L. Lopez, R. Papasin, 
C. Schulbach^ J. Van 

Software tools have been 
developed that accelerate the 
development of Grand Challenge 
Computational Aeroscience applica- 
tions for massively parallel process- 
ing systems. These software tools 
and associated research focus on 
areas critical to understanding and 
using computer systems capable of 



trillions of floating point operations 
per second, including: (1) perfor- 
mance tuning, visualization, and 
analysis; (2) parallel debugging; and 
(3) programming paradigms and 
environments. 

In the first area, an integrated 
performance-tuning environment is 
developed around the Automated 
Instrumentation and Monitoring 
System (AIMS). (See the first figure.) 
AIMS is a software tool that auto- 
matically instruments parallel 
FORTRAN and C programs, moni- 
tors their execution on parallel 
architectures, and displays and 
analyzes their performance. 
Dynamic variation of system param- 
eters and overall performance are 
automatically captured during 
execution and subsequently dis- 
played on the color screen of a 
workstation. A key feature of AIMS is 
that it provides source-code click- 
back, allowing the user to relate 
trace events with specific statements 
in the application program. AIMS 
also supports tracking and visualiza- 
tion of interprocessor data move- 
ments. In addition to providing trace 
information, AIMS has been used to 
investigate the possibility of regener- 
ating program trace information from 
statistics collected during program 
execution. 

The Numerical Aerodynamic 
Simulation trace visualizer (NTV) is a 
related tool that provides an alterna- 
tive to the visualization kernel (VK) 
of AIMS. Whereas VK provides 
dynamic visualization of programs, 
NTV provides static, zoomable 
views. Another product, Ben, is a 
library of visualization routines that 
can be used to develop visualization 
capabilities in other software tools 



R !■: V L r T I \ a k v T k c h n o l o t; v L h a i> s / I una rati re Tech not o^y aud Tools 



63 



Performance Tuning with AiMS 

Grand challenge High performance computer 

application 




Automated 

bottleneck 

detection 



Fig, 1. AIMS software toolkit for automatically instrumenting parallel 
programs, monitoring their execution, and displaying and analyzing their 
performance. 



and products. AIMS is available for 
the IBM SP-2 (message passing 
interface (MPI)) and on workstation 
clusters (parallel vertical machine 
(PVM) and MPI). NTV supports AIMS 
traces as well as native traces from 
thelBMSP-2. 

The second area involves the 
development of a debugger for use 
on parallel and distributed computer 
systems. The portable parallel/ 
distributed debugger (p2d2) (see the 
second figure) is a system using a 
client/server model that is portable 
across target machines and commu- 
nication libraries. This system allows 
the architecture- and operating 
system-dependent code to be 
located in a server while the client- 
side code remains highly portable. 
It was designed with scalable user 



Fig, 2. p2d2 being used 
to debug the running 
program mp_overflow. 




64 



R K V I. r T I N A R Y T E c H N I. G Y L F. APS / Ittnovative Technology and Tools 



Aerosavtics a s d Space Trassportatios 
t e c h s l g y enterprise 



interface elements in expectation 
that users will want to debug 
computations involving many (1 6 to 
256) processes. p2cl2 is currently 
available on the IBM SP-2 (for MPI). 
Development of support for PVM 
has also been completed. 

In the third area, programming 
paradigms and environments, the 
Cooperative Data Sharing System 
(CDS) has been developed to 
provide an efficient, portable, and 
simple system for creating processes 
and communicating between them 
in cluster, message-passing, and 
shared-memory environments. This 
setup allows the user to specify the 
semantics required of each commu- 
nication so that it can execute with 
near-optimal performance on any 
target architecture: shared-memory, 
message-passing, or cluster. The 
CDS application user interface 
consists of two layers: a kernel 
level (CDSl), with objectives of 
minimality, orthogonality, portabil- 
ity, efficiency, and utility; and a user 
level (CDS2) built upon CDSl , with 
the sole objective of providing a 
useful user interface without interfer- 
ing with the desirable characteristics 
of CDS1 . CDSl is available on SUN 
Solaris and Silicon Graphics Inc. 
Challenge Array systems. 

Point of Contact: C. Schulbach 

(650)604-3180 

c$chulbach@mail.arc.nasa.gov 



Numerical Aerodynamic 
Simulation Parallel 
Benchmarks 2.2 

William Saphir, 

R. Van der Wijngaart, 

Alex Woo, Maurice Yarrow 

The Numerical Aerodynamic 
Simulation (NAS) Parallel Bench- 
marks 2.2 (NPB2.2) bring a new era 
of integrity to the performance 
claims of the vendors of high- 
performance parallel computers. 
NASA is an immediate beneficiary 
of these benchmark tools, as is the 
scientific and engineering commu- 
nity at large, since evaluation of 
high-performance computers can 
now be based on objective perfor- 
mance criteria rather than vendor 
hyperbole. The NPB2.2 also stand 
out as excellent source code models 
for the development of future 
parallel versions of simulation 
programs for the aerospace industry. 

NPB2.2 consist of source code 
versions of seven computational 
fluid dynamics applications and 
kernels that were originally given as 
''paper and penciT' specifications. In 
their original form, they gained wide 
acceptance as measures of parallel 
computer performance, but vendor 
implementations were machine- 
specific and proprietary. The current 
source code versions, however, have 
been specifically designed to run on 
modern parallel computers that 
support the commonly available 
message passing interface (MPI). 
They are available from the World 
Wide Web, and have been designed 
to be extremely easy to use. Data 
generated by these benchmarks are 
now available for a wide variety of 
parallel computers from numerous 



vendors, and are readily viewable at 
the NASANPB Website. 

New to NPB2.2 are the IS 
integer sort kernel and a vastly 
improved FT Fast Fourier Transform 
kernel. Unlike the other kernels that 
measure floating-point performance, 
the IS code does integer operations 
only and measures raw processor 
power. It also provides a strenuous 
test of the speed of the communica- 
tion network of a modern parallel 
computer. The new FT program, no 
longer limited to 128 processors, has 
had its floating-point performance 
improved. The three application 
benchmarks known as BT, LU, and 
SP solve equations of viscous flow, 
each with a different algorithm. 
The remaining two kernels solve a 
multigrid problem and a so-called 
''embarrassingly parallel" problem. 
The suite provides a comprehensive 
test of the capabilities and relative 
performance level of a parallel 
computer, making the task of 
assessing high-performance comput- 
ers for NASA laboratories easier than 
ever before. 

Point of Contact: M. Yarrow 
(650) 604-5708 
yarrow@nas.nasa.gov 



Large-Scale Parallel 
Semiconductor Simulation 

Subhash Saini 

At the present time, several 
technologies are pressing the 
limits of the semiconductor-based 
computer industry, including the 
0.1 -micron foreseeable limit of the 
photolithography process, quantum 



K K V I. V T 1 N A R V T \i c w \ !. (I Y L F A p s / hui }' (i t i I e Tccbuology and Tools 



65 




1 Pflop/s 

Normal evolution 
Conventional supercomputers Tflop/s 



1 Gflop/s 



J I 



1990 1995 2000 2005 

Year peak performance first reached 



2010 



Strategy for implementing parallel 
adaptive unstructured-grid technol- 
ogy for semiconductor device 
modeling has been developed. This 
strategy addresses the important 
issues, such as load balancing. Also 
studied were the scaling properties 
of several algorithms, such as fast 
multipole expansions (FME) to 
determine the trade-off between FME 
and long-range molecular dynamics 
methods in terms of computational 
cost. Both these methods are used in 
device modeling. 

Point of Contact: S. Saini 
(650) 604-4343 
$$aini@mail.arc.nasa.gov 



Fig. 1. High-performance computing technology roadmap. 



effects, heat dissipation, the skyrock- 
eting cost of manufacturing plants, 
and others. NASA requirements for 
high-performance computers are 
different from those of mainstream 
industry. For many applications, 
such as the High-Performance 
Computing and Communication 
(HPCC) remote exploration and 
experimentation (REE) project, NASA 
requires low-powered, ultracompact, 
high-performance computers that are 
resistant to radiation damage. By the 
year 2010, many important future 
NASA missions will rely on petaflops 
computer systems. Applications of 
interest to NASA that will most likely 
require petaflops computing include 
multidisciplinary optimization of 
commercial aircraft designs, propul- 
sion systems, large-eddy and the 
direct numerical simulation of 
turbulence; computation of the radar 
scatter signature of aircraft bodies; 
full-scale simulation of the Earth's 



atmosphere and climate; and the 
design, manufacture, and operation 
of smart nanorobots for space 
exploration. 

A roadmap for the high- 
performance computing industry has 
been developed and is presented in 
the figure. Current consensus is that 
petaflops computing systems will be 
feasible in 20 years using advanced 
semiconductor technology. Ames 
Research Center has instituted a new 
program in semiconductor process 
and device modeling to ensure that 
NASA will be able to meet future 
requirements of petaflop computing. 
In spite of the seemingly obvious 
need for highly parallel computing 
as a simulation and design tool in 
this arena, so far any utilization of 
parallel systems has been rare. The 
principal impediments include 
access to parallel testbeds, as well as 
the availability of usable software 
and parallel computing expertise. A 




Nonlinear Aerodynamic 
Shape Optimization of 
High-Speed Research 
Configurations 

Susan Cliff, James Reuther, Ray Hicks 

The overall goal of this research 
is to develop an aerodynamic shape 
optimization technique that is based 
on nonlinear computational fluid 
dynamics (CFD) methods and 
control theory techniques; eliminates 
the dependence on a large number 
of CFD solutions; and can be 
effectively used to optimize aero- 
dynamic shapes to yield superior 
aerodynamic performance. In 
support of the high-speed research 
(HSR) program, the specific objec- 
tive is to apply these methods to 



66 



R E V L ij T 1 N A R Y T B c H N L G Y L H A p s / Intiovative Technology and Tools 



A E R S A i T I C S A S D S P A C E T R A S S P R T A T I .V 
T E C tt S n I (, \ E \ T E R P R } S E 



candidate configurations to assess 
their capabilities and simultaneously 
enhance the aerodynamic perfor- 
mance of the vehicle. 

The aerodynamic design for 
supersonic cruise of high-speed 
commercial transport aircraft is an 
extremely difficult problem because 
of the complex flow fields and the 
need to efficiently integrate the 
major components of the vehicle 
(i.e., the fuselage, wing, and engine 
nacelles) while satisfying numerous 
structural and operational con- 
straints. Past design efforts have 
relied upon linear theory methods. 
Unfortunately, the near-field 
interactions of the components are 
nonlinear and the use of linear 
theory-based design methods 
provides only limited design guid- 
ance. Furthermore, the level of 
cruise aerodynamic performance 
achieval)!e with linear design 
methods is insufficient to ensure an 
economically viable configuration. 

The role of CFD in the aero- 
space industry today is primarily 
that of an analysis tool used in the 
final stages of the design [process. 
Although this tool has proved very 
beneficial, to make full use of the 
power of CTt) it must be incorf)o- 
rated into the framework of an 
automated design process. Past 
et'forts to accomplish this objective 
have proved extremely difficult for 
applications involving complex 
aircraft configurations because of the 
com[)utational expense associated 
with the thousands of CTD solutions 
that are required to gain an under- 
standing of aerodynamic design 
sf)ace and its related optima. 

Traditional aerodynamic shape 
optimization methods use finite 



differences to obtain the necessary 
gradient of an aerodynamic objec- 
tive function (for example, drag) with 
respect to the design variables. Thus, 
if N is the number of design vari- 
ables used for the shape design, then 
N flow solutions are required to 
obtain the gradient during each 
design cycle. Through the use of 
control theory-based methods, a set 
of adjoint equations are formulated 
that reduce the computational 
requirement down to the equivalent 
of two flow solutions, independent 
of the number of control variables. 
The result is a computational savings 
of greater than an order of magni- 
tude. Furthermore, since the gradient 
calculations are no longer depen- 
dent on the number of design 
variables, the shape of complex 
configurations can be optimized 
using a large number of design 
variables. 

Using the nonlinear design/ 
o[)timization code named SYN87 
and its predecessors OPT'^D and 
OPTb7, which combine Euler flow 
solvers with constrained and uncon- 
strained numerical optimization 
algorithms, five successful aero- 
dynamic designs have been accom- 
plished. The first four designs have 
been validated through wind tunnel 
testing. The most recent application 
of this automated aerodynamic 
design f)rocess, which used SYN87 
and the adjoint gradient computa- 
tions, was focused on the Mach 2.4 
F^SR Technology Concept Airplane 
(TCA). The cruise aerodynamic 
shape of this configuration (shown in 
the figure (see Color Plate 14 in the 
Appendix)) was optimized to ol^tain 
the minimum drag subjec t to a large 
spectrum of realistic geometry 



constraints, including wing spar 
depth, wing leading edge bluntness, 
landing gear box size and location, 
fuel volume, passenger cabin 
dimensions, and floor attitude. The 
optimization procedure used up to 
140 independent design variables 
distributed over the wing and 
fuselage and focused on wing 
thickness and camber and fuselage 
camber and area distribution. Over 
200 design iterations were executed 
in the overall design process, and 
they yielded very significant perfor- 
manc e gains relative to the baseline. 
These perlormance gains were 
validated through numerous detailed 
Euler and Navier-Stokes analyses. 
Hardware is currently being fal)ri- 
cated for a wind-tunnel test to 
exf)erimenlally validate the results. 

This design/optimization 
approach represents some of the first 
successful attempts to use the [Knver 
of CFD as a true design tool applied 
to com[)lex aircraft configurations 
with a large realistic spectrum of 
constraints. In adciilion, the large 
performance gains achieved c learly 
demonstrate the advantage of 
using nonlincMr design methods to 
enhance aerodynamic fierformance. 

Point of Contact: D. Bencze 

(650)604-6618 

dbencze@maiLarc.nasa.gov 



R K \ L i T j N A R \ T !■; c li N I, ij (, Y L I- \ I' s / S !/ /> i> I's }i I c Tecbuol()\>y 



67 



Surface Operations 
Behavioral Evaluation 
Interim Testbed for 
High-Speed Research 

Mary K. Kaiser 

The external Vision System 
(XVS) element of the high-speed 
research (HSR) program is examining 
issues associated with building a 
high-speed civil transport (HSCT) 
with no forward-looking windows. 
Forward visibility would be provided 
by a synthetic vision system, gener- 
ally conceived as displaying visible 
light camera imagery with superim- 
posed symbology. This design 
concept presents numerous engi- 
neering and human factors chal- 
lenges, including the adequacy of 
the XVS display for control, naviga- 
tion, and obstacle detections- 
integration of the XVS display with 
the side windows (which are opti- 
cal); and crew coordination in an 
XVS-equipped flight deck. 

Surface operations may present 
particular challenges to an XVS- 
equipped aircraft because vehicle 
control is primarily vision-based 
(compared to many phases of flight 
in which the pilot relies heavily on 
instrument information) and because 
artifacts such as sensor-offset 
parallax become more severe when 
the aircraft is close to the ground. 

To examine these surface 
operations issues in a high-fidelity 
testbed, the HSR program is con- 
structing a near-full-scale testbed, 
the Surface Operations Research 
Evaluation Vehicle (SOREV). The 
SOREV is scheduled to be opera- 
tional in late FY97. in the interim, 




Fig. 1. The SOREV Externally mounted caniems feed the forward and 
inboard displays at the rear driver station. An outboard side window is 
configured to simulate the side window visibility planned for the HSCT. 



Ames has worked with industry 
partners at Boeing and Honeywell to 
equip a full-size Ford van (shown in 
the figure) with an XVS system 
located at a rear driver station. The 
rear station has two XVS displays 
(forward and inboard) and an optical 
window with the visibility angles 
specified by the Reference H design 
of the HSCT. The cameras can be 
mounted at several locations on the 
vehicle, creating parallax similar to 
that anticipated with aircraft 
mounted camera systems. Transport 
delays in the camera/display system 
can be introduced, and the impact 
on driver control and situational 
awareness can be examined. 
Although its overall fidelity is 
somewhat limited, the SOREV is 
proving to be a useful testbed for 



''first looks" at critical human factors 
issues associated with the XVS and 
surface operations. 

Point of Contact: M. Kaiser 
(650) 604-4448 
mkaiser@maiLarc.nasa.gov 



68 



R I- V L [■ T I N A R Y T [■ c II N L ci V L K A i> s / Supersofiic Tecljuoloii^y 



A h R ) \ t 11 ( S A S I) S }' .\ i i i k \ \ \ P it fi ! \ } ! \ 

T i- ( ii \ / u ) E \ I I K p R I s t: 




Propulsion Checkout and 
Control System 

Ann Patterson-Hine 

The Propulsion Checkout and 
Control System (PCCS) was part of 
the Integrated Propulsion Technol- 
ogy Demonstrator (IPTD) task 
supported by the X-33 program. 
The ol)jective of the IPTD was to 
predict and evaluate potential 
improvements in ground and flight 
propulsion-system operations in 
efforts to reduce turnaround times 
and to cut cost-risk fac:tors. The IPTD 
consists of liquid-oxygen and liquid- 
hydrogen propulsion modules, a 
remote interface unit that is a 
prototype of Rockwell's X-33 
Phase II propulsion system avionics- 
to-control-center interface, and a 
prototype control center at Marshall 
Space Flight Center's (MSFC's) 
Advanced Engine Test Facility. 

The task, which was completed 
in FY96, achieved the following: 
(1 ) on -board diagnostic/prognostic 
algorithms for the propulsion system 
of the Reusable Launch Vehicle 
were demonstrated to be capable of 
trending and analyzing a historical 
database; (2) playback of IPTD data 
on valve actuations identified early 
performance degradation of the 
12-inch prevalve in the main 
propulsion system that otherwise 
would have been undetectable; 
(3) the diagnostic/prognostic algo- 
rithms are the first known applica- 
tion of predictive Integrated Vehicle 



Health Management to an opera- 
tional cryogenic testbed; and 
(4) testability models developed by 
using a software tool set from 
QualTech Systems, Inc. were used in 
positioning the IPTD sensors. 

The design models were auto- 
matically converted to real-time 
code that monitored sensor data and 
that worked with the Langley smart- 
sensor routines to identify off- 
nominal conditions to the operator. 
This successfully demonstrated 
Rockwell's approach to reducing 
softw^are development costs by 
reusing design models and software 
for operational vehicle code. 

Point of Contact: A. Patterson-Hine 

(650)604-4178 

apatterson-hine@maiLarc.nasa.gov 



R K \ I I T I N A K \ T !■; C M N L (. V L i; \ I' S / ACCCSS (() S [) {{ C C 



69 



Space Science Enterprise 












S P A ( F S f I t \ ( E E y T E R P R { S E 



Overview 



NASA's Space Science Enter- 
prise seeks answers to fundamental 
questions about the origin o( the 
Universe, the origin and evolution ot 
galaxies and planetary systems, the 
origin and distribution of life in the 
Universe, and the connections 
between the Sun, Earth, and helio- 
sphere. Through our primary mission 
in astrobiology (the study of the 
living Universe), Ames provides 
pioneering research, technology 
development, and flight projects that 
promote fundamental discoveries 
about the origin, evolution, and 
distribution of life within the context 
of cosmic processes. Research is 
conducted to: 

• Determine the abundance and 
distribution of biogenic com- 
pounds that are conducive to 
the origin of life 

• Identify locations in the Solar 
System where conditions 
conducive to life have existed 

• Survey and begin surface 
explorations of the most fasci- 
nating and accessible planetary 
bodies 

• Discover the origin, evolution, 
and fate of the Universe, galax- 
ies, stars, and planets 

• Survey cosmic rays and interstel- 
lar gas as samples of extrasolar 
matter 

• Explore the frontiers of the 
heliosphere 

• Search for planets and planetary 
formation around other stars 



Exobiology and Astrobiology 

Ames has been recognized as a 
world leader in astrobiology and 
exobiology (search for extraterrestrial 



life) since the inception of these 
disciplines. Research at Ames 
focuses on revealing how life begins, 
on what processes govern its evolu- 
tion, on what destiny may be 
realized, and on how often life may 
have emerged on other planets. 
When coupled with Ames' pioneer- 
ing research on the dynamics of 
galaxies, molecular gases and 
clouds, planetary systems, and the 
Solar System, our study of life is 
facilitated by understanding the 
cosmic environment within which 
life evolves. Ames conducts studies 
of biologically important materials in 
geologic samples and in extrasolar 
matter; provides theoretical insights 
for addressing nv^jor issues in life's 
evolution; plays a key role as 
NASA's expert in matters of plan- 
etary protection; and helps to focus 
the science for future Mars missions 
on issues related to the possibility of 
the independent origin of life on 
Mars. 

One of the most important 
questions in exobiology and astrol^i- 
ology is how living systems emerge 
from molecular chaos. Highlighted 
in this report are exobiological 
simulations that revealed several 
new and important basic principles 
governing the ability of pefMides to 
acquire ordered structures, to self- 
catalyze, and to give rise to a simple 
mechanism for transmitting signals 
from the external environment to the 
interior of a protocell. 

The role of comets in the origin 
of life is also featured. Comets, 
which can be considered the fossils 
of the protoplanetary nebular 
processes, retain within their icy 
nuclei clues about the beginning of 
the Solar System. Ames reported an 
important finding this year on the 



71 



composition and properties of 
interstellar and connetary ices that 
may have profound implications for 
our understanding of the origin 
of life. 

Planetology 

Ames conceived and initiated 
7 of the 1 1 planetary missions that 
have flown since 1972, including 
the Mars Pathfinder. The Galileo 
probe made history on December 7, 
1 995, w^ith its entry into the atmo- 
sphere of Jupiter, the first time a 
spacecraft entered the atmosphere 
of an outer planet. Plunging into 
Jupiter's swirling storm clouds at 
106,000 miles per hour, the probe 
descended 400 miles through 
turbulence, violent winds, and 
clouds, transmitting data for almost 
an hour before the communications 
system succumbed to the intense 
Jovian environment. Results from the 
probe's descent are reported below. 

Ames also recently completed a 
comprehensive study of the Solar 
System which fully models the entire 
range of bombardment, transport, 
and mixing effects by which highly 
absorbing ''primitive" interplanetary 
material pollutes, and redistributes 
material in, an initially pristine icy 
system such as Saturn's rings. 
Surprising findings are described in 
this report. 

Of all the planets in the Solar 
System, Mars is most like Earth, 
especially in that its weather and 
climate arc dynamically similar to 
Earth's and in that its history may 
have included life. Research high- 
lighted in this report includes a 
new finding on the structure of the 
Martian regolith and its role in the 
Martian climate; the results of Ames 
research on the nature of the Martian 



oxidants— a primary obstacle to 
uncovering the biological history of 
Mars; and a unique concept, the 
Pascal Mission, to use Mars as a 
natural laboratory for the study of 
factors that control a planet's climate 
system. 

Astrophysics 

As NASA's lead Center in 
airborne astronomy, Ames pioneered 
and developed astrophysics and has 
the world's only capability for 
making astronomical observations 
from the stratosphere. Development 
was begun this year on the Strato- 
spheric Observatory for Infrared 
Astronomy (SOFIA), the infrared 
telescope of which will gather 
almost an order of magnitude more 
light than its predecessor and 
achieve a five-order-of-magnitude 
improvement in capability. Ames 
astronomers, who also contribute 
directly to space-based infrared 
science, won the right to the most 
observation time on the premier 
infrared space mission of the 
decade — the European Infrared 
Space Observatory. These and the 
highlights described beiow are 
presented more fully in later sections 
of this report. 

Ames and the University of 
Tokyo produced the Mid-Infrared 
Spectrometer on the Japanese 
Infrared Telescope mission which 
obtained infrared spectra of more 
than 10,000 objects and has yielded 
interesting new information on the 
character of the interstellar material. 
Results obtained this year confirm 
that two types of interstellar mate- 
rial — grains and certain polycyclic 
aromatic hydrocarbons (PAHs)— are 
mixed in a constant ratio throughout 
different interstellar regions of the 



Galaxy, hinting at common evolu- 
tionary processes. A complementary 
Ames study on PAHs revealed that a 
new type of PAHs, hydrogen PAHs 
(H-PAHs), found at the boundary of 
the Orion Nebula and Orion 
Molecular Cloud, are absent in 
higher-radiation environments. 
Results presented in this report 
indicate that different chemically 
active regions in planetary nebulae 
and molecular clouds are defined — 
at least in part — by the characteris- 
tics of the radiation environment. 

Protoplanetary nebulae, the 
birthplaces of planets and planetary 
systems, are distributed throughout 
the Universe. The "primary accre- 
tion" process — the process by which 
the first sizeable objects (comets, 
asteroids, etc.) formed from solids 
that entered a given protoplanetary 
nebula as micron-sized dust motes — 
is one of the major unsolved prob- 
lems of the origin of planetary 
systems. A new insight, achieved 
this year upon completion of a new 
model of turbulence and damping 
by particles in the midplane of a 
protoplanetary nebula, challenges 
certain theories of accretion 
dynamics. 

Ames researchers recently 
completed the most comprehensive 
theoretical computation of the 
potential energy surface and dipole 
moment for hot water in sunspots 
of the Sun and in the spectra of 
selected cool stars, revealing that 
sunspot atmospheres are much 
cooler in the shallow layers than had 
been predicted by previous models. 
These results and other potential 
applications of the model are 
highlighted in this report. 

The final stages in the life of a 
star somewhat more massive than 



72 



Space S c i t .\ r f E s t e r p r i 



s t 



the Sun was studied by Ames 
researc hers and colleagues from 
the University of Hawaii and the 
University of Wyoming. The results 
provide an important insight into 
understanding how the structural 
evolution of the nebula, through 
interacting winds, might provide the 
conditions necessary for molecular 
survival. 

Also described in this issue is a 
remarkable ''hands-on" educational 
software package developed by 
Ames that allows students of all ages 
to perform simulated observations of 
planets around other stars, as if they 
were the astronomers, and then to 
analyze the data to determine the 
habitable properties of the planets. 
Included in the package is a descrip- 
tion of Ames' proposed Kepler 
mission, the only currently practical 
method for finding Earth-sized 
planets around other stars. 

Space Technologies 

To support the Space Science 
Enterprise, Ames scientists and 
engineers develop, validate, and 
verify enabling, cutting-edge, high- 
payoff technologies for conducting 
future space science and exploration 
missions at substantially reduced 
costs. Cost reduction is achieved 
principally by using information 
technology to develop advanced 
flight- and ground-system autonomy 
technologies. The following high- 
lights are reported more fully in later 
sections. 

An executive system for intelli- 
gent autonomous spacecraft, 
planned for the Deep Space One 
mission, was prototyped this year. 
Deep Space One will autonomously 
navigate to asteroids and comets and 
will be able to achieve its mission in 



the face of a wide variety of faults, 
including total loss of command and 
control from Earth. 

Live tests of a key component of 
the Automatic Telescope Project 
were successfully completed. The 
primary objective is to provide 
advanced scheduling and automa- 
tion infrastructure so that the 
benefits and performance of single- 
user automatic telescopes can be 
made available to a large commu- 
nity of users, with no additional 
people in the loop. The benefits will 
be a dramatic reduction in telescope 
operating costs and an increase in 
both scientific productivity and ease 
of use. 

Ames' Assisted Space System 
Experimental Testbed (ASSET) 
project is an Internet-based global 
spacecraft control network that 
provides operational support of a 
variety of low-cost missions and a 
high-risk, low-inertia testbed for 
validating spacecraft autonomy 
strategies. This year ASSET enabled 
rapid experimentation with revolu- 
tionary autonomy technologies in a 
real-world, complex space system 
in support of the New Millennium 
program concerning the data 
distribution, principal investigator 
interfacing, beacon operations, 
ground-system automation, and fault 
management. 

Ames has developed WaveLab, 
a powerful suite of wavelet algo- 
rithms that can be used to character- 
ize the physical processes 
underlying complicated temporal 
and spatial variations, removing 
unwanted observational noise from 
data, compressing data with insig- 
nificant loss of information, and 
probing computer models for 
physical processes with large 



dynamic range. WaveLab is avail- 
able at no cost to the public on a 
World Wide Web site maintained at 
Stanford University. 

Significant progress has been 
made in understanding and improv- 
ing the operation of twt) of the main 
components of a pulse-tube 
cryocooler for use in spacecraft. In 
addition, Ames developed a star 
sensor that can be mounted in a 
cryogenic vacuum vessel to serve as 
a guide star-tracker for the Gravity 
Probe B relativity mission. 



73 



Remote Analysis of 
Martian Surface Materials 

D. Blake, P. Sarrazin, D. Bish, 

D. Vaniman, S. Chipera, S. A. Collins, 

T. Elliott 

A miniaturized x-ray diffraction 
and x-ray fluorescence (XRD/XRF) 
instrument for exploring the Mars 
surface is being developed; the 
instrument will minimize the 
uncertainties inherent in remote 
mineraiogical and geochemical 
analyses. The instrument has been 
named CHEMIN, in reference to its 
role as a simultaneous chemical and 
mineraiogical analyzer. A function- 
ing prototype of this instrument uses 
a copper X-ray tube, transmission 
geometry, and an energy-selective 
charge-coupled device (CCD) 
operating in single-photon counting 
mode to collect simultaneous XRF 
and XRD data. 

The ultimate use of this instru- 
ment vv^ill be in obtaining combined 
XRD and XRF data from planetary 
samples. Such an approach will 
greatly improve the accuracy of 
remote petrological analyses by 
constraining the number of possible 
mineralogic interpretations of the 
data. For example, although the 
Viking landers provided very useful 
XRF data on the Martian regolith, the 
complex chemistry reported (partic u- 
larly the mixed-anion suite that 
includes sulfur and chlorine, as well 
as oxygen) allows a wide range of 
mineralogic interpretations. Since 
remote XRF has yielded proven 
results, the focus is on preliminary 
tests in obtaining quantitative 



XRD data from the prototype 
CFHEMIN instrument. 

X-ray diffraction is the most 
direct and accurate analytical 
method for determining the presence 
of mineral species, because the data 
obtained by this method are funda- 
mentally linked to crystal structure. 
Other methods, based on chemical 
or spectral properties, are derivative 
and subject to much greater uncer- 
tainties. Moreover, significant 
progress has been made in the last 
decade in the development of 
quantitative XRD of multicom 
ponent mixtures. 



The Rietveld method, which fits 
the entire observed diffraction 
pattern with a pattern calculated 
using the crystal structures of the 
constituent phases of the model 
mineral system, shows great promise 
for mineral analysis. This method 
can provide rapid quantitative 
estimates of mineral abundance, as 
well as compositional, unit-cell 
parameter and structural data on 
individual minerals. In addition, 
recent advances in quadratic goal 
programming allow solution of 
simultaneous linear equations using 




Fi^. h Two-dimensional image of basalt diffraction pattern obtained with 
the CHEMIN instrument using only monoenergetic copper-characteristic 
radiation. The rings correspond to lattice spacings within individual minerals 
making up the basalt. 



74 



P K <; R !■; s s 



X 15 I f> L (. V 



Space S c i e s c f E \ r e r p r i s e 



chemical and mineralogic data. Use 
of these techniques with combined 
bulk-sample XRD and XRF data 
makes possible extraction of consid- 
erable information on individual 
mineral compositions. 

To test the capabilities of 
CHEMIN against realistically com- 
plex samples, Rietveld analysis was 
used to determine mineral abun- 
dances from CHEMIN-derived 
diffraction patterns of a basalt. This 
sample was chosen as representative 
of a rock type that is known to be 
abundant on the Martian surface, 
and also one that fits the generic 
mineralogy of all of the Mars 
meteorites identified so far (includ- 
ing the ALH84001 meteorite pur- 
ported to contain evidence of 
primitive Martian bacteria). 

XRD and XRF data were 
obtained from a sample of terrestrial 
basalt with the prototype CHEMIN 
instrument using repetitive 
30-second counts. After accumula- 
tion of numerous 30-second-count 
datasets, the data were analyzed on 
a pixel-by-pixel basis. X-ray powder 
diffraction rings were generated by 



3200 
2800 
2400 
2000 



I 1600 
1200 



800 



400 



Observed pattern 
Calculated pattern 




Difference pattern 

J \ I \ i L_ 



Background 



-LU I i I i L_ 



J I L 



12 



16 



20 



24 28 32 

2e ( Cu) 



36 



40 



44 



48 



Fif^. 2. Rietveld refinement of the diff!\)cto^rc\m obtained by summinf^ x-ray 
intensities shown in Fi^. 1 radi.illy around the centra! beam. The patterns are 
evaluated by least-squares fitting between model and actual results. 



plotting the two-dimensional 
distribution of those pixels contain- 
ing copper-characteristic radiation 
(first figure). The powder rings were 
then integrated to generate conven- 
tional 20-versus-intensity powder- 
diffraction patterns (29 is the angle 
between the diffracted beam and the 
entering beam). The resulting XRD 
data were used as input to Rietveld 
refinements. The results of the 
analysis are shown in the table and 
in the second figure. Despite the 
complexity of the basalt sample and 
the significant limitations of the 
prototype CHEMIN instrument, the 
Rietveld analysis is surprisingly 
good, with a calculated pattern that 



agrees well with the observed 
pattern. In addition, the resulting 
mineral analysis agrees well with 
optically determined modes for this 
sample. This level of success has 
only recently been obtained, and 
considerable improvement in the 
CHEMIN results are anticipated in 
the near future. 

The CHEMIN instrument has 
been proposed for a number of 
space missions, and an XRD/XRF 
capability is included in instrument 
suites proposed for Mars landers and 
rovers by several working groups. 

Point of Contact: D, Blake 

(650)604-4816 

dblake@mail.arc.nasa.gov 



P \< s> *i K i: s s [ s ]'] \ iM 1. (r V 



75 



stable Isotope 
Biogeochemistry of 
Hydrothermal Systems 

David J. Des Marais 

Hydrothermal systems probably 
existed in any rocky planet which 
was sufficiently large to accumulate 
substantial internal heat from 
radioactive decay and which also 
possessed a significant crustal 
inventory of water. Hydrothermal 
systems provide the liquid water, 
nutrients, reduced conditions, and 
chemical energy necessary to 
support life. Therefore NASA's 
astrobiology and solar system 
exploration programs should include 
a search for hydrothermal systems in 
other planets. 



The stable isotopic composition 
of the elements oxygen (O), hydro- 
gen (H), sulfur (S), and carbon (C) in 
minerals and other chemical species 
can indicate the existence, extent, 
conditions, and processes (including 
biological activity) of hydrothermal 
systems. The reactions between hot 
water and rock-forming minerals 
allow the isotopic ratios ^^^Op^'O of 
deuterium to hydrogen (D/H) of the 
minerals to equilibrate with this 
water, causing isotopic changes 
which can be used to detect fossil 
systems and to delineate their area! 
extent. These isotopic changes can 
indicate fluid composition, alteration 
temperatures, water-rock ratios, and 
so forth. The ^^^O/'^^O values of 
hydrothermal silica and carbonate 



Biosynthesis 




cc 


.,: 








Sea, Atm. 






1 


' Decompos 


i 
tion 


L 






Fresh organic matter 






Marine 
HCO3" 












DecofT 


iDOsltion and burial 






Burial 




Sedimentary organic matter - 


Weathering 


-^— 


Carbon- 
ates 




1 ► 


i Out- 
gassing 


















Hydrothermal 
reduced carbon 


Hydrothermal 
CO2 and carbonate 


^ ■ 
















4 4 


4 










Subduction 


LJ 

► 


Mantle 
















car 


bon 


S 


ubdt 


ction 



-40 



-30 



-20 



-10 



+10 



5'^C 



Fig. /. Carbon isotopic composition ot crustal carbon reservoirs, with 
special emphasis on hydrothermal systems. The horizontal axis depicts 
variations in ^ 'C/^'C values in parts per thousand, with lower values to the 
left Boxes and arrows depict carbon reservoirs and processes linking the 
reservoirs, respectively The isotopic composition of hydrothermal carbon 
species can be affected both by mantle carbon and by contamination from 
crustal sources of organic and carbonate carbon. 



deposits tend to increase with 
declining temperature and thus 
help to map temperature gradients 
within hydrothermal systems. The 
^^C/'^'C values of carbonates and 
' t-depleted microbial organic 
carbon increase along the thermal 
gradients of thermal spring outflows, 
principally because of the outgassing 
of relatively ^ ^C-depleted carbon 
dioxide (CO2). Hydrothermal waters 
can differ in their D/H contents 
because of their variety of sources, 
which can include seawater or 
rainwater, which falls over ranges in 
elevation and geographic location. 
Therefore, measurements of D/H 
values in hydrothermal minerals can 
help to decipher the origins of the 
fluids. The ^V^S and ^\:P'C 
values of fluids and minerals reflect 
the origin of the S and C, the partial 
pressure of O2, and also the nature 
of key oxidation-reduction pro- 
cesses. For example, a wide range 
of ^"^S/^^S values is consistent with 
equilibration below lOO^C between 
sulfide- and sulfate-bearing minerals, 
and can be attributed to sulfur 
metabolizing bacteria and nofto 
nonbiological processes. Depending 
on its magnitude, the difference in 
the ^^C/^^C value between carbon- 
ate minerals and coexisting organic 
carbon can be attributed either to 
equilibrium at hydrothermal tem- 
peratures or, if the difference in the 
ratio exceeds 1 %, to organic biosyn- 
thesis (see figure). 

The observations of hydrother- 
mal mineral assemblages and 
coexisting organic carbon in 
Yellowstone National Park indicate 
that both temperature gradients and 
microbial communities can indeed 
be documented through measure- 
ments of their oxygen and carbon 



76 



P R (i K K S S IN K X h i L (; V 



Space S c i t: .\ ( t E \ t f r p r i s t: 



isotopic compositions. Parallel 
measurements are now being 
made in rocks from an ancient 
(400-mill ion-year-old) hydrothermal 
system in Queensland, Australia. 

Point of Contact: D. Des Marais 
(650) 604-3220 
ddesmarais@mail.arc.nasa.gov 



Fossilization Processes in 
Thermal Springs 

lack Farmer^ Sherry Cady, 
David ). Des Marais 

Stromatolites are laminated 
biose^dimentary structures that form 
when communities of microorgan- 
isms either entrap sedimentary 
grains, or induce the precipitation of 
minerals through their metabolic 
processes. Fossil stromatolites 
provide a primary source of informa- 
tion about the early biosphere on 
Earth, and have also been identified 
as imfK)rtant targets in the explora- 
tion tor a fossil record on Mars. 
These megascopic biosedimentary 
structures are rich storehouses of 
paleobiological information and 
often contain not only the fossilized 
remains of the microorganisms 
which produced them, but also 
biochemical and isotopic signatures 
that directly link them to life 
processes. 

Thermal springs are highly 
productive microbial factories that 
provide an important glim[)se into 
the nature of the primitive biosphere. 
In recent years, the microbial 
communities and stromatolites of 
thermal springs have assumed a 
prominent place in studies of 
primitive biospheres. During the 



early history of Earth, it is likely that 
hydrothermal systems were wide- 
spread owing to higher internal heat 
flow, and to the thermal energy 
invested by large impacts. Thus, 
hydrothermal environments are 
regarded as key environments for the 
origin and early evolution of terres- 
trial life, a view that also receives 
support from the universal tree of life 
derived by comparing the nucleic 
acid base sequences in the RNA of 
living organisms. The patterns of 
early branching suggest that the 
common ancestor of all living 
organisms was an extreme thermo- 
phile, that is, a microorganism that 
preferred living at temperatures 
above 80"C(176^F). 

The fossil record of thermal- 
spring deposits is virtually unknown, 
and no subaeriai systems have yet 
been recognized in the ancient 
geological record of Earth. In order 



to create a framework for recogniz- 
ing such deposits in the Precambrian 
record, the authors have been 
carrying out studies of modern 
thermal-spring systems In 
Yellowstone National Park, along 
with ancient deposits from localities 
in western North America and 
Australia. 

In FY96 studies of modern 
thermal springs led to several 
discoveries that have helped refine 
the understanding of stromatolite 
morphogenesis and fossilization 
processes at high temperatures. 
Columnar geyserites (see first figure), 
are high-temperature subaeriai 
splash deposits that form around the 
vents of siliceous thermal springs. 
The structures thus formed resemble 
many Precambrian stromatolites. 
Columnar geyserites were first 
studied in the late 1970s and 
determined to be abiogenic (that is, 




F/^>. /. Thin-section view oi columnar stromatolites (geyserites) formed 
around high-temperature vents of silica-depositing springs in Yellowstone 
National Park; columns are about 3 mm high. 



i\ \ iM (1 1. (i (i V 



77 




Fig. 2. Thin-section view of fossilized microbes from 350-million-year-old 
siliceous thermal-spring deposits discovered in the Drummond Basin ot 
northeast Queensland, Australia; individual filaments are about 20 fum in 
diameter. 



not produced by living organisms). 
Since that time, some authors have 
questioned the biogenicity of the 
oldest fossil stromatolites, suggesting 
that they were also formed inorgani- 
cally. But studies of modern geyser- 
ites have shown conclusively that 
these deposits are actually formed in 
the presence of thin biofilms of 
thermophylic bacteria and archae- 
bacteria, and that the geyserite 
microstructure reflects a significant 
biogenic contribution. 

In FY96 the first detailed paleon- 
tological studies of ancient subaerial 
thermal-spring deposits in northeast 
Queensland, Australia, were also 
completed. These 350-million-year- 
old deposits are precise analogs for 
the modern subaerial systems that 
have been studied in Yellowstone. 



Observations were made of the 
primary biofabrics and microfossils 
that are preserved in these ancient 
deposits, despite extensive textural 
reorganization during diagenesis 
(second figure). 

Point of Contact: J. Farmer 
(650) 604-5748 
jfarmer@mail.arc.nasa.gov 



Molecular Biomarkers for 

Stromatolite-Building 

Cyanobacteria 

Linda L. Jahnke, Roger E. Summons, 
Harold P. Klein 

Life began on early Earth in an 
environment devoid of free molecu- 
lar oxygen (O2). It is clear that the 
rise of O2 as a consequence of the 
evolution of oxygenic photosynthesis 
had a profound effect not only on 
the subsequent progression of 
biological evolution but also on 
Earth's geological history. Although 
oxygenic photosynthesis evolved 
within the cyanobacterial lineage, 
evidence for this event's timing 
remains obscured within the 
molecular and rock records. One 
potential help in resolving this issue 
resides in the complex organic 
molecules that can be extracted from 
ancient sedimentary rocks. These 
molecules, called biomarkers, are 
the remnants or molecular fossils of 
ancient microorganisms. The carbon 
skeletons of such fossil molecules 
can survive structurally intact for 
billions of years, and organic 
geochemists are working to identify 
them in ancient organic sedimentary 
materials. The challenge is to 
understand the link between such 
molecular fossils and their 'living" 
counterparts, the biomarker mol- 
ecules synthesized by contemporary 
bacterial groups. Understanding the 
function and influence of environ- 
mental factors on the synthesis of 
such biomarker molecules in the 
source organism can then provide a 
basis for interpretation of this ancient 
organic record. 



78 



P K (i K !■: s s IN K X !^ I 1. (; v 



Space S c i e s c f. E \ t e r f r i s e 



^' ^ 



'r/ 



yy /- 






F/X^ /. Photomicro^mph of d Phormidium (filament width 1,5 jjm) isoLited 
from ci Yellowstone microbial mat. 



Extant cyanobacteria are a 
diverse group of prokaryotic micro- 
organisms found in a wide variety of 
modern environments. We have 
focused our study on the types of 
cyanobacteria involved in the 
construction of microbial mats and 
their fossilized equivalents, stroma- 
tolites. These laminated, organo- 
sedimentary structures are among 
the most common fossils in the 
geological record. Important 
biomarkers for identification of 
cyanobacterial input to sedimentary 
organic matter are a group of 
straight-chain hydrocarbons 
(alkanes) having 1 7 to 20 carbons 
(Ci7-C2()) with midchain, 
branched methyl groups such as 
7-methylheptadecane (C17). These 
methylalkanes are specific to 



cyano[)acteria, are resistant to 
biodegradation, and form some of 
the oldest known biomarkers, having 
been isolated from Proterozoic rocks 
dated at 1.7 billion years ago. More 
recently, we have identified another 
class of methylated alkane having 
two methyl branches at carbons 7,1 1 
or 7,10 in a cyanobacterium, 
Phormidium luridum. We found that 
these dimethylheptadecanes (DMA) 
were only synthesized when the 
cyanobacterium was grown with low 
levels of dissolved inorganic carbon 
(Die:). Cells grown with DIC levels 
equivalent to atmospheric levels of 
carbon dioxide (CO^) (0.0.3'K)) 
contain DMA, whereas cells grown 
with 17<> CO2 have only a straight- 
chain n-heptadecane. We have 
carried out a screening of a variety 



of cyanobacteria to determine the 
extent to which this low DIC-DMA 
synthesis phenomenon occurs. To 
date, we have screened 6 marine 
and 1 2 freshwater strains of 
cyanobacteria. None of the marine 
cyanobacteria, including several 
Synechococcus, three Oscillatoria, 
and a Pseudanabaena were able to 
synthesize DMA. This was also true 
for a variety of cyanobacteria 
( Cloeocapsa, Synechococcus, 
Oscillatoria, Pseudanabaena) 
isolated from the hot spring, micro- 
bial mats in Yellowstone National 
Park. Two other Yellowstone 
isolates, Chlorogloeopsis and 
Fischerella, did synthesize 
dimethylalkanes, but these com- 
pounds differed in carbon chain 
length and methyl group positions, 
and were synthesized under both 
high and low DIC conditions. The 
only cyanobacteria found to be 
capable of dimetylheptadecane 
synthesis were four Phormidium 
types isolated from siliceous hot 
springs (see first figure). Like 
Phormidium luridum, the 
Yellowstone Phormidium contained 
high levels of DMA when grown 
under low DIC conditions. 

The Phormidium are responsible 
for the formation of a type of colum- 
nar stromatolite, the Conophytons 
(conical-formed microfossils). 
Conophytons were one of the most 
distinctive groups of stromatolites 
which disappeared near the end of 
the Precambrian. The Pliormidium 
mats in Yellowstone provide a 
surviving "primitive" analog for 



\< <; \{ !■: s s I \ 



\\ \ I) w i 



79 




f/,^^ 2. Silicifying Phormidium mat (width 12 inches) showing Conophyton 
structure. Weeds Pool Yellowstone National Park. 



study of the Conophytons (second 
figure). DMA are found only in 
Precambrian oils, and disappear 
from younger organic sediments just 
as the Conophyton stromatolites 
disappear from the rock record. The 
environmental factors controlling 
dimethylalkane synthesis in the 



Yellowstone Phormidium mats may 
provide important insights for 
interpretation of the organic molecu- 
lar fossil record. 

Point of Contact: L. Jahnke 
(650) 604-3221 
ljahnke@mail.arc.nasa.gov 



Earth-Threatening Comets 
Leave Tell-Tale Dust Trails 

Peter jenniskens, David Morrison 

Over its geologic history, Earth 
has been bombarded by a continu- 
ous shower of large asteroids and 
comets, some of which led to large- 
scale extinctions. Unlike other 
dangers, impacts of larger bodies 
can affect mankind as a whole and 
pose a threat to modern civilization. 
The first step toward avoiding such 
impacts is to know where the 
potential impacting objects are. Most 
large asteroids can be found by 
dedicated searches using large 
telescopes, because they have a 
short orbital period and come close 
to Earth frequently enough to be 
detected. EHowever, that is not the 
case for comets with long orbital 
periods, which come back to the 
inner parts of the solar system only 
once every 200-1 00,000 years. And 
some 2 to M)% of all such impacts 
are thought to be due to these long- 
period comets. 

Researchers are studying the 
possibility of using the meteoroid 
particles in the path of these comets 
as a means of detecting their pres- 
enc:e. Erom numerous accounts of 
amateur observers, they know that 
some meteor streams show briet 
episodes of high activity: the meteor 
outbursts. At such times, Earth 
crosses relatively recent ejecta of 
particles, which evaporate in Earth's 
atmosphere in the flashes of light 
that are called meteors. Most 
intriguing are the outbursts that 
occur at times unrelated to the return 
of the comet to perihelion, and it is 



80 



P R (1 \{ [■: s s IN K \ 1^ 1 I, f) (i V 



S P A ( I S f / / \ ( f E \ r t R P R I S E 



suspected that these might be 
associated with long-period comets, 
it remained unclear whether the 
outbursts were caused by a trail of 
dust in the path of these comets or 
by clouds of dust in orbits with a 
shorter orbital period. No observa- 
tions made with standard meteor 
observing techniques had been 
successful, until FY96. 

It was proposed that the plan- 
etary motions affect the orbits of 
particles, causing Earth to periodi- 
cally traverse the trail of dust. The 
return of one such stream in Novem- 
ber of 1995 was predicted. Condi- 
tions would be favorable to measure 
the orbital period of the particles 
over Europe and Africa, where it was 
night at the expected time of the 
event, and the point from which the 
meteors radiated would be high in 
the sky, allowing good viewing. A 
network of photographic and image- 
intensified TV systems was set up in 
Andalusia, Spain, operated princi- 
pally by amateur observers of the 
Dutch Meteor Society. 

The outburst did indeed occur 
and was confirmed by many observ- 
ers in Europe. Meteor rates increased 
briefly to five meteors per minute. 
Most of the meteors were too weak 
to be photographed. A rare color 
image is shown in the figure (see 
Color Plate 1 5 in the Appendix). Ten 
meteors were recorded from mul- 
tiple sites and triangulation of their 
paths allowed the reconstruction of 
the orbit of the particles. The 
particles had a long orbital period, 
confirming that this meteor outburst 
was due to a trail of dust in the wake 
of a long-period comet. Identifica- 
tion was made of 14 such streams 
that produce occasional outbursts 



and techniques are being developed 
to find more in order to study their 
properties. 

Point of Contact: P. )enniskens 
(650) 604-3086 
peter@max.arc.nasa.gov 



Capturing Cosmic Dust 
on Mir 

Kenji Nishioka, Ted Bunch, 
Mark Fonda, Glenn Carle, 
Sherwood Chang, James Ryder, 
Janet Borg 

As part of exobiology's effort to 
solve the quintessential question of 
how life started on Earth, it is 
necessary to have information 
concerning the prehistoric origin and 
evolution of biogenic elements and 
compounds from the interstellar 
medium that coalesced into the solar 
system. Interplanetary and cosmic 
dust particles (IDPs/CDPs) are 
believed to have survived from that 
early period largely unchanged; as a 
result, they may shed light on the 
chemical pathways taken by the 
biogenic elements and their com- 
pounds in moving from their origins 
in stars and the surrounding media 
to their incorporation into planetary 
bodies such as Earth. Thus the 
interest to capture contamination- 
free cosmic dust particles. 

Ames Research Center, the SETI 
Institute, and Lockheed Martin 
Missiles and Space have developed 
ultrapure aerogel, containing about 
one part per million carbon, for 
contamination-free capture of CDPs 
in Earth orbit. This group joined with 
the University of Paris's Institute for 



Astrophysical Studies and flew a 
CDP-capture experiment early this 
year (October 199S to February 
1996) as part of the European 
Exposure Facility's Comrade experi- 
ment on the Russian Mir space 
station. Two small capture cells/ 
modules each 9 by 5 by 1 .5 centi- 
meters thick were flown, recovered, 
and returned to Ames on March 2S, 
199b. One of the two aerogel 
ca[)ture modules was exposed from 
early October 1995 until early 
February 1996; the second capture 
module was exposed for only 
10 days during the Orionids shower 
in October 1995. These capture cells 
were the first aerogel experiments to 
fly on Mir. The results of preliminary 
analyses of the capture modules are 
summarized below. Strict [proce- 
dures were followed in handling the 
modules to ensure that any changes 
detected, including contamination, 
would be a result of the environment 
around Mir, from shipping of the 
modules (packing material), or from 
captured particles. 

Physical (visual) examinations 
of the modules were done in a 
10,000 class clean room. When the 
module covers were removed, the 
integrity of the module/aerogel was 
verified; it had survived shipping, 
launch, orbital exposure, and 
recovery without apparent structural 
problems, except for stretching of 
the platinum retainer wires. This 
probably resulted when entrained air 
in the aerogel could not escape 
quickly enough to equalize the 
internal pressure and the rapidly 
decreasing external pressure during 
launch, thus causing the aerogel to 
bulge and the wires to stretch. The 
cosmetic surface blemishes on the 



P K (; K K S S ! \ K \ 1^ 1 1. (} (i V 



81 



aerogel surface reported in the 1995 
Research and Technology Report, 
appeared noticeably darker and 
rougher, whereas the clear unblem- 
ished surfaces appeared unaffected 
in the recovered modules; this was 
later confirmed by microscopic 
examination. These initially blem- 
ished areas are probably susceptible 
to ozone oxidation. 

While in the clean room, surface 
and subsurface aerogel samples 
were taken from the modules for 
analyses. Organic chemical analyses 
of the aerogel surface samples 
showed high contamination levels 
(2%) of dioctyl adipate, a plasticizer. 
This was unanticipated, and there 
are no reports of this contaminant as 
an orbital experiment concern in the 
literature. Additional samplings from 
the modules confirmed the high 
contamination levels. Interior 
samples showed that the contamina- 
tion did not migrate and was con- 
fined to the surface. 

An automated microscopic 
scanning system at the University of 
California at Berkeley was used to 
scan the modules for impacts. 
Several candidate impact points with 
tracks were found and are being 
analyzed by an x-ray probe tech- 
nique at Brookhaven National 
Laboratory. Preliminary x-ray 
fluorescence spectra results from the 
x-ray probe of a track failed to detect 
particulate matter at the end of the 
track. But, the "chondritic" elements 
calcium (Ca), iron (Fe), nickel (Ni), 
manganese (Mn), titanium (Ti), and 
chromium (Cr) were found in the 
track and could indicate the breakup 
of a pyroxene particle; however, 
small amounts of lead (Pb), zinc 
(Zn), copper (Cu), strontium (Sr), and 
zirconium (Zr) were also found, 



strongly suggesting the possibility of 
an orbital debris particle or other 
source of contamination. 

Valuable insights were gained 
into the manufacturing, fabrication, 
and use of aerogel for building a 
successful IDP/CDP capture instru- 
ment from this flight experiment. 
Aerogel is robust and holds together 
even under IDP/CDP impact, and 
under launch and ground handling 
loads. Also, aerogel does not appear 
susceptible to the migration of 
surface contaminants. But as 
expected, the experiments gave rise 
to new questions. For example, 
where did the high concentration 
of dioctyl adipate come from? Do 
particles, especially at the micron or 
tens of micron size, survive capture, 
and how will these micron-size 
particles and their tracks be found 
and analyzed cost effectively? 

Point of Contact: K. Nishioka 

(650)604-0103 

knishioka@mail.arc.nasa.gov 



Simple Peptides at 
Water-Membrane 
Interfaces 

Andrew Pohorille, Christophe Chipot 

Even the simplest protocell must 
have had the capability to catalyze 
the chemical reactions needed for its 
survival and growth and to commu- 
nicate with its environment. These 
functions must have been accom- 
plished by simple molecules that 
could have been present in a 
protobiological milieu. One such 
group of potential early catalysts and 
signaling molecules were the 



peptides — possible precursors of 
enzymes and receptors. Unfortu- 
nately, short peptides typically have 
disordered structures in aqueous 
solution and, therefore, do not 
appear to be suitable for the desired 
cellular functions. However, at 
water-membrane, water-oil, or 
water-air interfaces, many of these 
peptides, depending on their 
sequence, can acquire a broad range 
of well-defined secondary structures, 
such as a-helix, b-strand, or b-turn. 
A crucial, common characteristic of 
these interfaces is that a nonpolar 
phase is adjacent to water. 

The ability of small peptides to 
organize at aqueous interfaces was 
examined by performing a series of 
large-scale molecular dynamics 
computer simulations of several 
peptides. The peptides were com- 
posed of two amino acids, nonpoiar 
leucine (L) and polar giutamine (Q), 
in a variety of environments; they 
differed in the size and sequence of 
the amino acids. Among the mol- 
ecules studied were the dipeptides 
LL, LQ, QL, and QQ. Although these 
peptides were too short to form a 
secondary structure, they are very 
good models for use in examining 
the conformational preferences of 
the peptide backbone as a function 
of the sequence and the environ- 
ment. The studies of L/Q peptides 
were extended to include two 
heptamers, of sequence LQQLLQL 
and LQLQLQL. The sequences were 
designed to maximize the interfacial 
stability of the a-helix and b-strand 
conformations, respectively, by 
exposing polar side chains to water 
and nonpolar side chains to a 
nonpolar phase. 

Finally, a transition of an 
undecamer (1 1 residues), composed 



82 



P \i (i \{ \: s s IN H \ H I I. (1 V 



Space S c i e s c e E s r e r p r i s e 



entirely of leucine residues, from a 
disordered structure in water to an 
a-helix in a nonpolar phase, repre- 
senting the interior of the membrane, 
was investigated. This is the first time 
that the complete folding of a 
peptide in solution was accom- 
plished in computer simulations. 

The simulations revealed several 
basic principles governing the 
sequence-dependent organization of 
peptides at interfaces. Understanding 
these principles allows for determin- 
ing how peptides could have 
performed protocellular functions. 

Short peptides tend to accumu- 
late at interfaces and acquire 
ordered structures, provided that 
they have a proper sequence of 
polar and nonpolar amino acids. 
The specific identity of amino acids 
appears to be less important, which 
is a desirable protobiological 
property. The dominant factor 
determining the interfacial structure 
of peptides is the hydrophobic effect, 
which is manifested at aqueous 
interfaces as a tendency for polar 
and nonpolar groups of the solute to 
segregate into the aqueous and 
nonpolar phases, respectively. The 
resulting structures are called 
amphiphilic. Exceptions from this 
tendency are observed only when 
intramolecular hydrogen bonds are 
formed in the nonpolar phase 
without completely removing polar 
side chains from water, a unique 
feature of interfacial systems. 

Based on the results for folding 
the LQQLLQL heptamer to an 
a-helix at the water surface, it is 
proposed that, whenever possible, 
peptides fold at interfaces through a 
series of amphiphilic intermediates. 
Transitions between these intermedi- 
ates involve changes not only in 



backbone angles but also in confor- 
mations of side chains. Once folded, 
the peptides form structures that are 
suitable for polymerization and have 
the potential for catalytic activity. 
If peptides consist of nonpolar 
residues only, they insert into the 



nonpolar phase. As demonstrated by 
the example of the leucine 
undecamer, such peptides fold into 
an a-helix as they partition into the 
nonpolar medium (see the two 
figures). The folding proceeds 




Fig. /. Polyleucine undecamer interacting with the water-hexane interface 
att = nanoseconds. The center of mass of the peptide is located 
approximately 11 angstroms from the water-hexane interface. In water, the 
structure of the peptide is disordered. The atoms of the polyleucine are dark 
gray the oxygen and hydrogen atoms of water are medium gray and white, 
respectively and the CH^ and CH^ segments of hexane are light gray 



P ^ (; H K s s IN Ex B I I, (; 



83 




Fig. 2, Polyleucine undecamer interacting with the water-hexane interface 
at t = 40.2 nanoseconds. The peptide has folded into an a-helix, and the 
peptide backbone vector makes an angle of about 20° with the normal to 
the water-hexane interface. The color scheme is the same as described 
in Fig. 1 . 



through an intermediate conforma- 
tion, called a 3io-helix, which 
remains in equilibrium with the 
a-helix. Once in the nonpolar 
environment, the peptides can 
readily change their orientation with 
respect to the interface from parallel 



to perpendicular, especially in 
response to local electric fields. 
The ability of nonpolar peptides to 
modify both their structure and 
orientation with changing external 
conditions may have provided a 



simple mechanism for transmitting 
signals from the environment to the 
interior of a protocell. 

Point of Contact; A, Pohorille 
(650) 604-5759 
apohorille@mail.arc.nasa.gov 



Silicon-Micromachined 
Gas Chromatography 
System 

Thomas Shen, James Suminto, 
Frank Yang, Daniel Kojiro, 
Glenn Carle 

The development of gas chro- 
matographic equipment for plan- 
etary atmospheric probes and soil 
gas or pyrolytic analysis was re- 
quired for both the Viking mission to 
Mars and the Pioneer Venus mission. 
A gas chromatography (GC) instru- 
ment was used for the Pioneer Venus 
mission. However, future missions, 
such as the Mars Exploration Mis- 
sions and missions to probe the 
atmospheres or moons of the outer 
planet, will require further reduc- 
tions in equipment weight, volume, 
and power requirements. 

NASA is in a very dynamic 
mode at this time. A massive effort is 
under way to revolutionize the 
technical approach to spacecraft and 
robotic exploration. The Discovery 
Program is the first of these new 
approaches. Other programs follow- 
ing it include New Millennium, 
Micro-spacecraft, and Nanoinstru- 
ments. All of these enterprises have 
in common the requirement for 
better, faster, cheaper. Clearly, 
current technology will not likely be 
feasible for future space studies. 



84 



P R (} G K K S S IN E \ B I [, C Y 



Space S c i e s c e E s t e r p r i s e 



.04 



.03 



< 
£ 



.02 



3 

o 



.01 




Detector #6 
Detector #5 
Detector #4 



100 




200 300 

Voltage (V) 



400 



500 



Fig. h Discharge curves of micromachined chips. 



A silicon-micromachined GC 
system in which the detector and 
column are integrated into a small 
piece of silicon has been studied by 
several researchers. The problem is 
that no sensitive detector has been 
designed and tested. A proposal was 
made to use the metastable ioniza- 
tion detector principle, and replace 
the radioactive material with the 
glow discharge method, which has 
not been studied on the silicon- 
micromachined GC system. A very 
narrow discharge distance can be 
produced and controlled, a helium 
ionization detector can be produced, 
and a highly sensitive analysis can 
be obtained (concentrations below 
the parts-per-million level can be 
detected) through a micromachined 
silicon wafer. In addition, a column 



on the chip with the new in-house- 
developed highly efficient silicon 
polymer material (U.S. patented) will 
be modified and coated to separate 
most chemical components. This 
small and highly sensitive GC system 
will be able to meet future mission 
requirements. 

In this first year of the project, 
several micromachined chips with 
different electrode designs were 
prepared and tested. The preliminary 
devices were produced by Micro- 
Tech Scientific Inc. Several chips 
have been evaluated. The discharge 
curves of these chips are demon- 
strated in the figure. 

Point of Contact: T. Shen 
(650) 604-5769 
tshen@maiLarc.nasa.gov 



Nitrogen Sources and 
Sinks on Early Earth 

David P. Summers 

One important form of fixed and 
reduced nitrogen on early Earth is 
ammonia. However, current 
geochemical evidence points to a 
nonreducing atmosphere which 
contains nitrogen {N2) instead of 
ammonia. One source of ammonia 
involves the formation of nitrite 
(NO3") which can then be reduced 
to ammonia by ferrous iron (Fe'*'^). 
It has also been suggested that this 
cannot be an important source of 
ammonia since nitrite can react with 
ammonia (to make nitrogen). 

Whether this is true, however, 
depends on how the kinetics of the 
reactions involved compare with the 
speed of other sources and sinks for 
both ammonia and nitrite. If the 
reaction of nitrite with ammonia is 
slow compared with other processes 
that make and destroy ammonia, 
then it will not be important. Any 
evaluation of the importance of 
various sources and sinks of nitrite 
and ammonia, and of whether 
ammonia is likely to be present in 
significant amounts, requires an 
analysis of all the kinetic processes 
affecting those species. 

Using a kinetic analysis of rate 
data that had either been measured 
at Ames or were already published 
in the literature, a kinetic analysis 
was performed to establish what the 
concentrations and rates might have 
been in the early ocean. Plausible 
sinks/sources for nitrite and ammo- 
nia on early Earth are; the reduction 
of nitrite to ammonia, the reaction of 
nitrite with ammonia, photochemi- 
cal destruction of either nitrite or 



P K (I K K S S I 



-X H I I (; V 



85 



ammonia, and destruction at hydro- 
thermal vents. The photochemical 
destruction of nitrite depends on 
trace species that can react with 
photo-generated hydroxy radicals 
and on the amount of ultraviolet 
light, neither of which is known for 
certain to have existed on early 
Earth. Furthermore, the product of 
photolysis, nitric oxide (NO), is the 
same compound that nitrite (and 
hence ammonia) is formed from. 
Therefore, values for rates of photo- 
chemical destruction probably 
represent only an upper limit. 
Ammonia can also be lost by 
adsorption into clays. However, this 
will be significant only when the 
ammonium concentration is high 
enough to compete with potassium 
and serves mostly to limit the 
concentration below an upper limit 
of 10 millimolar. 

Under most conditions, the 
primary sink for nitrite is reduction 
to ammonia. Destruction at hydro- 
thermal vents is important at acidic 
pH's and close to the freezing point 
of water. The reaction between 
ammonia and nitrite is not an 
important sink for either nitrite or 
ammonia. Photochemical destruc- 
tion, even in a worst case scenario, 
is unimportant under many condi- 
tions but could be important under 
acidic, high carbon dioxide (CO2) 
pressure, or low-temperature 
conditions. The primary sink for 
ammonia is photochemical destruc- 
tion in the atmosphere. Under acidic 
conditions, more of the ammonia is 
tied up as ammonium (reducing its 
vapor pressure), and hydrothermal 
destruction becomes more impor- 
tant. At 25°C, pH 7.6, and 0.2 atmo- 
sphere of CO2, nitrite and ammonia 
have steady-state concentrations of 



3x10^^ and 2x10"^ molar, respec- 
tively. The very low nitrite concen- 
tration reflects the fact that the nitrite 
is reduced to ammonia before it has 
a chance to build up or undergo any 
competing processes. The concen- 
tration is such that its importance for 
many processes envisioned for the 
origin of life (like the formation of 
amino acids) is unclear. 

This is part of a broader question 
of whether life would have started 
in the open ocean, where dilution 
makes the formation of more 
complex species difficult, or some- 
where where a means of concentra- 
tion species (such as with vesicles 
or by evaporation) is present. In 
general, concentration would tend to 
raise the concentration of all species, 
rather than change the relative 
importance of the source and sinks 
(except where species are not 
allowed to be photolyzed or swept 
into hydrothermal systems). 

Point of Contact: D. Summers 
(650) 604-6206 
dsummers@mail.arc.nasa.gov 



Planetary Rings 

Jeff Cuzzi 

Planetary ring and moon 
systems, especially the magnificent 
Saturn system, contain elements of 
processes in particle disks that 
provide insights into planetary 
formation. Saturn's rings will be a 
main focus of study for the Cassini 
mission, to be launched in October 
1997. Interplanetary meteoroids 



bombard planetary rings, generating 
ejecta which transport mass and 
angular momentum in systematic 
ways throughout the system. An 
exhaustive study, which fully models 
the entire range of bombardment, 
transport, and mixing effects by 
which highly absorbing ''primitive'' 
interplanetary material pollutes and 
redistributes material in an initially 
pristine icy system such as Saturn's 
rings, has been completed. The 
study includes careful treatment of 
the fate of ejecta particles as they fly 
about within a realistic radial profile 
of ring particle abundance, and also 
detailed radiative transfer modeling, 
using actual material refractive 
indices of the individual ring particle 
reflectivity as a function of wave- 
length and location in the rings. 

The model implies (1 ) that with 
most parameters of the model 
constrained by other observations, 
it is found that the basic global 
brightness and color differences 
between the major ring regions 
(A-ring, B-ring, C-ring, etc.) can be 
understood for the first time, and in a 
simple way, by differential darkening 
caused by the infall of spectrally 
neutral, highly absorptive, probably 
carbonaceous, meteoritic material 
with orbits much like those of 
comets; (2) that the ''initial," primor- 
dial composition of the rings is 
almost entirely water ice, but must 
contain a uniformly distributed, 
highly red, probably organic mate- 
rial such as is found in "primitive" 
bodies in the outer solar system and 
in Titan's organic atmospheric haze 
(silicates are not acceptable as 
coloring agents); (3) that the radial 
profile of ring particle color (and 
thus compositional) variations 
provides a good fit to the model 



86 



K {) (I K i: S S IN 



X (} H I {) L (; Y 



S P A C F S f ! E \ ( F, E S T E R P R f S E 



profiles for a range of exposure 
times which is in agreement with 
exposure-time estimates from 
bombardment effects on radial 
variations of ring structure; and 
(4) that nominal parameters for the 
model imply a ring age of the order 
of 100 million to 1 billion years, 
depending on the ejecta yield and 
the meteoroid flux, indicating that 
the rings may not be as old as the 
solar system (4.5 billion years). 

Point of Contact: J. Cuzzi 
(650) 604-6343 
jcuzzi@mail.arc.nasa.gov 



Planetesimal Formation in 
the Protoplanetary Nebula 

Jeff Cuzzi 

The ''primary accretion'' 
process — the process by which the 
first sizable objects (comets, aster- 
oids, etc.) formed from solids that 
entered the protoplanetary nebula as 
micron-sized dust motes — is one of 
the major unsolved problems in 
understanding the origin of planetary 
systems. One key accomplishment 
this year was the completion of a 
new model of self-consistent turbu- 
lence generation and damping, by 
thin, dense layers of large (10-cm 
radius and larger) particles in the 
midplane of a protoplanetary 
nebula. Prior studies neglected the 
role of the particles in damping 
turbulence in the gas; a scheme to 
account for this effect has been 
developed and implemented. The 
scheme begins with the Reynolds- 
averaged equation for the turbulent 
kinetic energy. The full two-phase 



equations are averaged, and new 
terms appear as a result of particle 
damping, which has been modeled 
using a gradient-diffusion-type 
approach. 

In FY96, several different 
parameterizations for dissipation of 
turbulence by the gas itself, which 
has a major effect, were compared. 
A two-length scale model, a law-of- 
the-wall model, and two variable 
eddy-frequency models were 
explored; few really significant 
differences in model behavior were 
found between these parameteriza- 
tions. Significant damping of nebula 
turbulence does occur, to a degree 
that is particle-size dependent, but 
the damping effect is not sufficiently 
strong to allow the particle layer to 
settle into a gravitationaliy unstable 
dense midplane layer as some prior 
workers had speculated. This 
appears to be one of the first fully 
coupled, physically parameterized 
studies of turbulence damping by 
mass loading — certainly this is so in 
the context of an orbiting disk. 

Three-dimensional direct 
numerical simulations of particles in 
turbulence have been exhaustively 
analyzed. The Taylor microscale 
Reynolds number ranges from 40 to 
1 40, and the code handles one 
million particles at each of 16 Stokes 
numbers (the Stokes number is the 
ratio of particle stopping time to 
Kolmogorov eddy turnover time). 
Previously, it was demonstrated that 
the particle size most strongly 
concentrated (by orders of magni- 
tude) is in good agreement with 
chondritic meteoritic data. This year, 
it has been demonstrated for the first 
time that the particle density field for 
the most strongly concentrated 
particle size is a multifractal — and 



that a certain key descriptor of the 
multifractal (the so-called singularity 
spectrum) is Reynolds-number 
independent. 

Quantitatively, the density-field 
singularity spectrum agrees with that 
for turbulent dissipation, perhaps 
providing some profound clues 
toward explaining this very impor- 
tant process. It is now feasible to 
make use of extensive mathematical 
techniques developed for fractal 
analysis to predict the probability 
density of particle concentration at 
arbitrary Reynolds numbers, 
strengthening the applicability of 
these results to plausible nebulae 
with Reynolds numbers that are 
orders of magnitude higher than are 
achievable numerically. 

Point of Contact: J. Cuzzi 
(650) 604-6343 
jcuzzi@mail.arc.nasa.gov 



PASCAL: A Mars Climate 
Network Mission 

Robert M. Haberle, David C. Catling, 
Steven C. Merrihew 

Of all the planets in the solar 
system, Mars is most like Earth. In 
particular, its weather and climate 
are dynamically similar to Earth's. 
Mars therefore offers a natural 
laboratory for the study of the factors 
that control a planet's climate 
system. Yet the kind of observations 
that are needed to carry out such a 
study, namely long-term simulta- 
neous measurements from a network 
of globally distributed surface 
stations, are lacking in NASA's Mars 
exploration program. To remedy 



P l< (; R K S S i N Pi. A N F T A K V S Y 



87 



this, a Mars climate network mission 
was proposed in response to the 5th 
Announcement of Opportunity to 
NASA's Discovery Program. This 
mission was named PASCAL after 
the 1 7th century French scientist 
whose name is synonymous with 
the Systeme International unit of 
pressure. 

Pressure is the key measurement 
of the PASCAL mission. It provides a 
measure of the total mass in a 
column of air. Given information on 
its horizontal and temporal variation 
one can infer the nature of the 
planet's general atmospheric circula- 
tion and climate system. Fortunately, 
pressure is also easy to measure, and 
pressure sensors do not require 
orientation or deployment. Thus, 
very simple landers can be designed 
that are light and robust. In effect, 
we are trading off the ability to make 
many different kinds of measure- 
ments at a few sites, for just a few 
measurements at many sites. This 
trade allows us to achieve the global 
coverage that is required to observe 
the planet's climate system while 
keeping the total mission cost well 
within the Discovery cap. 

PASCAL'S mission objective is to 
establish a network of 24 globally 
distributed surface stations that will 
measure hourly variations in pres- 
sure and temperature over a period 
of 10 Mars years (-18.8 Earth years). 
This long-term continuous presence 
will define the nature and variability 
of the planet's climate system on 
time scales ranging from hours to 
years. It will also provide a weather 
monitoring infrastructure for future 
Mars missions. In addition to 
network science, PASCAL will gather 
data on the structure of the atmo- 
sphere as each probe descends to 



the surface. This aspect of the 
mission will be particularly valuable 
for future Mars orbiters which plan 
to aerobrake into their final orbits. 

PASCAL would be launched in 
April of 2001 and arrive at Mars in 
January 2002. The 6.8-kilogram 
probes would be released from a 
spin-stabilized carrier on direct 
approach to Mars. To achieve global 
coverage, the probes would be 
released in salvos of eight, each 
separated by a propulsive time-of- 
arrival adjustment to allow Mars to 
rotate underneath. Entry descent and 
landing would utilize a heat shield, a 
parachute, and crushable material 
for final deceleration. One possible 
network configuration resulting from 
this deployment scheme is shown in 
the figure (see Color Plate 1 6 in the 
Appendix). 

To ensure long lifetimes, each of 
PASCAL'S surface stations would be 
powered by a Light Weight Radio- 
isotope Heating Unit (LWRHU) 
coupled to a thermoelectric con- 
verter. The LWRHU also provides 
thermal control for the surface 
stations. Communication would 
occur through one of the orbiters of 
the Mars Surveyor Program (MSP). 
MSP plans to have orbiters at Mars 
beginning in September 1997 and 
continuing well into the next 
century. As a safety precaution, 
however, PASCAL'S landers are 
capable of storing up to 5 Mars years 
of data should an orbiter not be 
available. 

Point of Contact: R. Haberle 
(650) 604-5491 
rhaberle@mail.arc.nasa.gov 



The Center for Star 
Formation 

D. Hollenbach, P. Cassen 

The Center for Star Formation 
Studies, a consortium of scientists 
from the Space Science Division at 
Ames and the astronomy depart- 
ments of the Universities of Califor- 
nia at Berkeley and Santa Cruz, 
conducts a coordinated program 
of theoretical research on star and 
planet formation. The Center 
supports postdoctoral fellows, senior 
visitors, and students; meets regu- 
larly at Ames to exchange ideas and 
to present informal seminars on 
current research; hosts visits of 
outside scientists; and conducts a 
week-long workshop on selected 
aspects of star and planet formation 
each summer. 

The Star Formation Summer 
Workshop in 1996 was held at 
Wellesley College in Wellesley, 
Massachusetts, and included, in 
addition to the Ames scientists, 
about 100 astrophysicists from all 
over the world. 

The main focus of work in FY96 
was on the collapse of dense clouds 
to form protostars with protoplane- 
tary disks, the evolution of these 
disks, which ultimately form planets, 
and the influence of young massive 
stars on the surrounding star-forming 
molecular cloud. D. Hollenbach, 
A. Tielens, and M. Kaufman mod- 
eled the thermal structure and 
spectra of dense molecular clouds 
which are collapsing to form high 
mass stars. A similar calculation was 
made by C. Ceccarelli (University of 
Grenoble), D. Hollenbach, and 
A. Tielens for clouds collapsing to 
form low-mass stars. When the 



88 



P R (; K H s s IN Plan li t a k y S y s t i- m s 



S p A ( h S ( 1 1: \ ( /: E \ r t: R p R I s t 



model spectra are compared with 
observations, observers can identify 
collapsing clouds and measure the 
rate at v^hich the protostellar mass is 
increasing. D. Hollenbach and 
D, Neufeld (Johns Hopkins Univer- 
sity) calculated the heating and 
ionization created when the collaps- 
ing cloud hits the protoplanetary 
disk. Comparisons of the models 
with recent radio observations of 
hot, ionized gas surrounding a 
nearby protostar made it possible to 
determine the rate of mass accretion 
onto the disk and the mass of the 
central protostar. K. Chick and 
P. Cassen developed models to 
explain the survival of presolar 
material during the formation of the 
solar system, as dust particles in the 
collapsing molecular cloud eventu- 
ally incorporate into primitive 
meteorites. 

Once formed in the collapse, 
protoplanetary disks around solar- 
type stars evolve for millions of years 
as they form planets. K. R. Bell and 
P. Cassen examined the shapes of 
protoplanetary disks and demon- 
strated how the temperature depen- 
dence of the effective grain opacity 
can lead to configurations in which 
much of the disk is shadowed from 
the central star. B. Pickett, P. Cassen, 
and R. Durisen (Indiana University) 
used three-dimensional hydrody- 
namic simulations to show how the 
development of gravitational Insta- 
bilities in the early stages of proto- 
planetary disks is controlled by the 
radiant energy losses. 

Young massive stars radiate 
copious amounts of ultraviolet 
radiation, which heats, dissociates, 
and ionizes the surrounding molecu- 
lar cloud, and which ultimately 
leads to the dispersal of the cloud 



and the cessation of star formation 
in the cloud. D. Hollenbach and 
H. Stoerzer modeled these pro- 
cesses, incorporating a large 
numerical code which followed 
the time-dependent chemistry, the 
thermal balance, and the transport of 
radiation. Their models predict the 
infrared spectra from such regions, 
so that observations can be inter- 
preted quantitatively in terms of the 
rates of these destructive processes. 
These theoretical models have 
been used to interpret observational 
data from such NASA facilities as the 
Infrared Telescope Facility, the 
Kuiper Airborne Observatory, and 
the Infrared Astronomical Observa- 
tory, as well as data from numerous 
ground-based radio and optical 
telescopes. In addition, they have 
been used to determine require- 
ments for future missions such as 
the Stratospheric Observatory for 
Infrared Astronomy and the pro- 
posed Space Infrared Telescope 
Facility. 

Point of Contact: D. Hollenbach 

(650)604-4164 

dhollenbach@maiLarc.nasa.gov 



Energetic Trapped 
Particles near Jupiter 

John D. Mihalov 

On December 7, 1995, the 
Galileo Probe carried experiments 
Into Jupiter's atmosphere, near the 
planet's equator. The experiment 
complement included an energetic- 
particle telescope supplied by a 
group at the University in Kiel, West 
Germany. Data from this experiment 
could improve earlier measurements 



of the innermost Jovian energetic 
trapped proton fluxes, made with the 
Pioneer 10 and 1 1 spacecraft, as 
well as provide data from the region 
even closer to the planet that had 
not yet been directly traversed by 
spacecraft. Beginning at 3 hours 
before atmospheric entry, inside the 
orbit of to, and until entry, the 
experiment sampled the energetic 
trapped particle environment. Flux 
measurements and energy spectral 
and angular distribution data for 
energetic electrons and protons were 
obtained. 

The electron fluxes appear to 
be within a factor of 3 of values 
obtained by earlier experiments 
aboard Pioneers 10 and 1 1, at 
locations where comparisons could 
be made. Except nearest to the 
atmosphere, where the energy 
spectrum steepened considerably, 
the most probable measured elec- 
tron energy was about 20 and 
40 million electron volts, respec- 
tively, for each of two separate 
electron-energy thresholds, based 
principally on energy-dependent 
responses for the experiment 
deduced from Monte Carlo simula- 
tions. Energetic-proton fluxes, 
spectra, and angular distributions 
were measured in three separate 
energy ranges in the 42 to 
--131 million electron-volt energy 
range. The proton fluxes were 
observed to decrease abruptly 
inward from the location of Jupiter's 
ring, and also as Jupiter's atmo- 
sphere was approached. 

Future analyses of the decreases 
in particle fluxes measured as the 
atmosphere was approached may 
give insight into the structure of 
Jupiter's near-equatorial upper 



V R (; K E S S I N P I A N V. \ A H V S Y S T 



89 



atmosphere, because the energetic 
particles lose energy as they interact 
with atmospheric constituents while 
they drift around the planet and 
bounce along the magnetic field. In 
addition, atmospheric interactions of 
the electrons are characterized by 
Coulomb scattering that causes 
diffusion of mirror points along 
magnetic field lines. Some energetic 
trapped helium, in the 62 to 
1 36 million electron-voit per 
nucleon energy range, evidently was 
detected outside Jupiter's ring. 

Point of Contact: J. Mihalov 

(650)604-5516 

jmihalov@mail.arc.nasa.gov 



Wavelet Software 

|eff Scargle 

Wavelets are a mathematical 
tool for the analyzing and interpret- 
ing time series, images, and other 
datasets. New wavelet-based 
methods, developed for scientific 
and engineering applications, are 
useful for the following: (1 ) Charac- 
terizing the physical processes 
underlying complicated temporal 
and spatial variations; (2) removing 
unwanted observational noise from 
data; (3) compressing data with only 
insignificant loss of information; 
(4) probing computer models for 
physical processes that have a large 
dynamic range. 

A powerful suite of wavelet 
algorithms to implement all these 
ideas has been developed and made 
publicly available under the name 
''WaveLab/' WaveLab was a coop- 
erative research project between 



NASA Ames Research Center and 
the Statistics Department of Stanford 
University through joint Research 
Interchanges, and was funded by the 
NASA Astrophysics Data Program 
and Astrophysics Software and 
Research Aids Program. The author 
collaborated with David Donoho 
of the Stanford and U.C. Berkeley 
Statistics Departments on this 
project. The actual software is in the 
form of MatLab (copyright the 
MathWorks, Inc.) scripts, which can 
be easily translated to other systems. 

The World Wide Web site, 
maintained at Stanford (http:// 
playfair.Stanford.EDUiSO/'-wavelab/) 
includes downloadable software for 
Unix, Macintosh, and Windows 
computers, complete documenta- 
tion, tutorials in basic and advanced 
wavelet transform methods, sample 
data, and browser routines for 
exploring the sample data. 

The WaveLab system has been 
downloaded by several thousand 
persons who have registered their 
use of this free, public-domain 
software, many of whom have 
indicated that they have already 
found WaveLab useful for their 
research, teaching, or commercial 
enterprises. The developers answer 
questions and help with user 
problems by e-mail. 

Point of Contact: J. Scargle 
(650) 604-6330 
jeffrey@sunshine.arc.nasa.gov 



Time-Dependent 
Structures in Galaxies 

Bruce F. Smith, Richard A. Gerber, 

Richard H. Miller, 

Thomas Y. Steiman-Cameron 

Numerical experiments on 
models of galaxies are yielding new 
insights into galactic structures and 
dynamics. This research is designed 
to examine time dependencies in the 
structural components of galaxies 
and to determine the implications for 
both the structural and dynamical 
evolution of galaxies. The results of 
these investigations are being used 
in interpreting and understanding 
high-resolution observations of 
galaxies made by the Hubble Space 
Telescope (HST) and other advanced 
observational programs. A series of 
numerical experiments has been 
made on the Cray C90 and parallel 
machines at the Numerical Aerody- 
namic Simulation (NAS) Facility at 
Ames Research Center to understand 
the nature of time-dependent 
processes in galaxies. These experi- 
ments probed the nature of complex 
motions observed at the centers of 
galaxies including those seen at the 
center of our Galaxy. 

Global oscillatory modes have 
also been seen in the numerical 
experiments. These n-body experi- 
ments provided the first experimen- 
tal evidence of long-lived global 
oscillatory motions in spherical 
systems. These oscillations have 
been identified as normal modes; 
the entire galaxy rings like a bell. 
Several modes have been identified, 
but two are dominant and both are 
spherically symmetrical. The first is 
nearly a homologous contraction 



90 



P IM) G R !-: S S IN P L A M K r A R V S Y S T 1; M S 



Space S c i f s c e E s r e r p r } s e 



and expansion of the entire galaxy 
(the fundamental mode; a "breath- 
ing" mode), and the second has one 
node. A detailed study of the effects 
of softening in the representation of 
the potentials in the calculations has 
verified the earlier results on long- 
lived global oscillations. 

Given that the normal-mode 
oscillations persist over many 
dynamical time scales, a natural 
question is: What processes could 
give rise to the oscillations? Among 
the possibilities are that (1) the 
oscillations could be left over from 
the galaxy formation process, 
(2) they may be initiated by inter- 
actions with neighboring galaxies, 
or (3) they could grow from 
instabilities. 

These global oscillatory modes 
may play an important role in the 
structure and evolution of disk/halo 
galactic systems. The approach has 
been to assume that a significant 
fraction of the mass in a galaxy is in 
a dark halo. The halo oscillates and 
the luminous disk material responds 
to these oscillations. In the numeri- 
cal experiments the disk responds 
to the time-varying potential by 
developing a ring structure that 
forms and disappears during each 
oscillation cycle (see first figure (see 
Color Plate 1 7 in the Appendix)). 
This pattern of response persists over 
time periods approaching a Hubble 
time (assumed age of the Universe 
since the big bang). The response 
can be understood as a resonance of 
the stars in an annular region where 
the disk epicycle frequency is the 
same as the halo oscillation fre- 
quency. This appearance of the ring, 
or annular gap, is reminiscent of 
some SO galaxies (disk-like struc- 
tures), such as NGC 451 3. Such 



systems are not uncommon and 
might be explained by the interac- 
tion of a disk with an oscillating 
halo. As the evolution proceeds a 
bar develops in the inner disk (see 
second figure (see Color Plate 18 
in the Appendix)). Many galaxies, 
including our own, have bars in their 
central regions. 

Point of Contact: B. Smith 

(650)604-5515 

bfsmith@mail.arc.nasa.gov 



Galileo Encounters 
Jupiter: Results from the 
Probe 

Richard E. Young 

The cone-shaped Galileo probe 
entered the atmosphere of Jupiter on 
December 7, 1 995, at a speed of 
over 106,000 miles per hour and 
survived deceleration forces of 
228 times Earth^s gravity. After 
deploying a parachute, it relayed 
data to the Galileo orbiter space- 
craft, which was overhead, for 
57.6 minutes. At the same time, the 
Galileo orbiter began a 2-year, 
1 1 -orbit tour of Jupiter. So far it has 
had close flybys of the Galilean 
moons Ganymede, Callisto, and 
Eurof)a. The images it provides have 
spatial resolutions 10-100 times 
better than the Voyager images. 

The Galileo probe project is 
managed by NASA Ames Research 
Center. Hughes Space and Commu- 
nications Co., El Segundo, Califor- 
nia, designed and built the probe. 
Lockheed Martin Hypersonic 
Systems (formerly General Electric), 



Philadelphia, Pennsylvania, built 
the probe's heat shield. NASA's jet 
Propulsion Laboratory, Pasadena, 
California, built the Galileo orbiter 
spacecraft and manages the overall 
mission. 

The ratio of helium to hydrogen 
by mass is a key piece of information 
in theories of planetary evolution. 
For the Sun, this value is about 257c>. 
Comprehensive analysis of results 
from the probe's helium-abundance 
detector have shown that the helium 
abundance for Jupiter is near 247n, 
about the value for the primordial 
Sun. 

The lovian helium al)undance 
indicates that gravitational settling of 
helium toward the interior of Jupiter 
has not occurred nearly as fast as it 
apparently has on Saturn, where the 
approximate helium-to-hydrogen 
ratio is just b%. This indicates that 
Jupiter is much hotter in its interior 
than its neighbor Saturn, the next 
largest planet in the solar system. It 
also may cause scientists to revise 
their estimates of the size of the 
rocky core believed to exist deep in 
the center of Jupiter. 

The abundances of several 
heavy elements, including carbon, 
nitrogen, and sulfur, are significantly 
greater on Jupiter than on the Sun. 
This implies that the influx of 
meteorites and other small bodies 
into Jupiter over the eons since its 
formation has played an important 
role in Jupiter's evolution. 

However, minimal complex 
organic compounds were detected, 
indicating that such complex 
combinations of carbon and hydro- 
gen are rare on Jupiter and that the 
chances of finding biological activity 
there are extremely remote. 



P K (t (. K !■: S S IN Pi. A \ !•: T A K V S V S T 



91 



The winds, determined from 
tracking of the probe, persisted deep 
below the clouds, strongly suggest- 
ing that heat escaping from deep in 
the planet's interior drives the winds, 
rather than solar heating. Since all 
the outer giant planets exhibit strong 
winds, scientists hope that under- 
standing Jupiter's winds will lead to 
important new insights Into their 
unusual meteorology. 

The Galileo probe apparently 
entered Jupiter's atmosphere near 
the southern edge of a so-called 
infrared hot spot, which is believed 
to be a region of reduced clouds. 
The probe's nephelometer (an 
instrument used to study suspended 
particles) observed only one distinct 
cloud layer, and that a tenuous one 
by Earth standards. It is likely to be 
an ammonium hydrosulfide cloud. 
Three distinct cloud layers (an upper 
layer of ammonia crystals, a middle 
layer of ammonium hydrosulfide, 
and a thick bottom layer of water 
and ice crystals) were expected. 
There were, however, some indica- 
tions from the probe net flux radiom- 
eter and nephelometer that the 
probe started taking direct measure- 
ments in the bottom part of Jupiter's 
top ammonia ice cloud layer. 

Further analysis of probe data 
has confirmed the preliminary report 
that the Jovian atmosphere appears 
to be relatively dry, with much less 
water than anticipated on the basis 
of solar composition and on predic- 
tions from data sent by the Voyager 
spacecraft that flew by Jupiter in 
1 979. These studies predicted a 
water abundance for the planet of 
twice the solar level (based on the 
Sun's oxygen content.) Actual probe 
measurements now suggest an 



amount of water much less than that 
of the Sun. 

The probe's instruments found 
much less lightning activity on 
Jupiter per unit area than on Earth. 
Lightning on Jupiter was found to be 
about one tenth of that found on 
Earth in an area of the same size. 
Although there is less lightning 
activity, the individual lightning 
events are, however, about 10 times 
more energetic than similar events 
on Earth. 

Point of Contact: R. Young 
(650) 604-5521 
reyoung@mail.arc.nasa.gov 



Regolith Effects on Mars' 
Climate 

Aaron Zent 

In FY96, analysis of the adsorp- 
tion of water on Martian analog 
materials at Mars-like conditions was 
completed. In laboratory experi- 
ments, it was discovered that the 
mass of adsorbed H2O was approxi- 
mately an order of magnitude less 
than previously predicted. These 
results were used to derive new 
adsorption isotherms which reflect 
the heterogeneity of the surface sites 
on Martian (or any) soil. 

In collaboration with Howard 
Houben at Ames, these new iso- 
therms were applied to the study of 
the transport of water by the Martian 
atmosphere. It was found that the 
horizontal advection of water by 
the Martian atmosphere could not 
reproduce the observed patterns of 
atmospheric H2O variations, but that 



substantial vertical exchange 
between the atmosphere and the 
regolith could explain the results. In 
collaboration with Bruce Jakosky at 
the University of Colorado, it was 
found that nighttime inflections in 
the atmospheric temperature record 
at the Martian surface could be 
explained by the condensation of 
H^O in the atmosphere, at a frost 
point which supports the model of 
diurnal H^O exchange. 

Point of Contact: A. Zent 

(650)604-5517 

azent@mail.arc,nasa.gov 



The Nature of the Martian 
Oxidants 

Aaron Zent 

In FY96, laboratory analysis of 
the ability of hydrogen peroxide 
(H2O2), complexed by the Ti4-h 
cation, was completed to reproduce 
the results of the Viking biology 
experiments. It was discovered that 
the results of the Labeled Release 
experiment, in which organic 
molecules were oxidized by a 
temperature-labile oxidant in the 
soil, could be reproduced by outer- 
sphere (that is, with an intervening 
hydroxyl radical (OH) group) 
peroxo-Ti complexes. The ability of 
inner-sphere (that is, no intervening 
OH group) complexed H2O2 to 
reproduce the Gas Exchange experi- 
mental reactivity and stability pattern 
is still being investigated. 

Also explored was the role of 
the diffusion of H2O2 into the 
Martian regolith, and the role of 



92 



H (I R !■: S S I N 



I A N !•: T ,\ R V S V S T V. \i S 



Space S ( i e ^ ( e E s t e r p r i s e 



impact gardening in redistributing 
oxidized material through the 
Martian regolith. One finds that for 
reactive half-lives of the order of 
100,000 years, diffusion v^ill carry 
the oxidant into the upper 20 meters 
of the Martian regolith; also impact 
cratering will stir this material 
throughout the upper 100 meters of 
the regolith, necessitating deep 
drilling if unoxidized material is to 
be recovered. 

Point of Contact: A. Zent 

(650)604-5517 

azent@mail.arc.nasa.gov 



A Thermo-Acoustic 
Oxidant Sensor 

Aaron Zent 

In FY96, an investigation was 
l)egun into the feasibility of using a 
thermo-acoustic oxidant sensor to 
identify and quantify the nature of 
atmospherically produced oxidants 
in the Martian atmosphere. A series 
of selective chemical coatings, 
including lead sulfide, metal phtha- 
locyanines, and f)eroxidase are 
placed on chemiresistors and surface 
acoustic-wave devices. The mode of 
measurement for the chemiresistors 
is resistivity change owing to 
bonding between the measure and 
the coating; for the acoustic-wave 
devices it is mass changes of the 
same coating. Sensitivity to hydro- 
gen peroxide, ozone, and carbon 
monoxide is demonstrated. 

In collaboration with Dr. Marc 
Madou of the University of Califor- 
nia at Berkeley, flight electronics for 
this experiment are being developed. 
In the coming year, the crosstalk and 



effect of background gases on the 
proposed array of coatings will be 
investigated. 

Point of Contact: A. Zent 

(650)604-5517 

azent@mail.arc.nasa.gov 



PROCiKfSS /,\ As 



Astrobiology in the 
Astrochemistry Laboratory 

Louis J. Allamandola, Scott Sandford, 
Max Bernstein, Robert Walker, 
Dave Deamer 

The Ames Astrochemistry 
Laboratory has made major contri- 
butions to the understanding of the 
composition and properties of 
interstellar and cometary ices. This is 
done by studying laboratory analogs 
of cometary and interstellar ices. An 
important application of this work to 
the origin of life is reported here. 
Since comets are considered to be 
a major source of the molecules 
brought onto primitive Earth before 
life began, their organic composition 
is of central importance to questions 
concerning the origin of life. 

Cosmic ices contain the very 
simple molecules, water (HiO), 
methanol (CHjOH), carlx)n monox- 
ide (CX)), carbon dioxide (CO2), 
molecular hydrogen (H2), ammonia 
(NH .), and formaldehyde (H.CO), 
as well as more com[)lex sfxn ies. 
Ultraviolet irradiation of these ices 
produces other simple species such 
as H., H.CO, CO2. CO, methane 
(CH4), formyl radical (HCO), and 
more complex organic molecules 
known as amides, nitriles, amines, 



ketones, aldehydes, and polymers 
which are of particular interest from 
a prebiotic perspective. The ready 
formation of these complex organic 
species from simple starting mixtures 
and the ice chemistry that ensues 
when these ices are warmed suggest 
that interstellar ices and comets may 
have played an important part in the 
origin ot life. 

In addition to cometary ices 
raining a rich inventory of poten- 
tially important prebiotic molecules 
down onto early Earth, the com- 
pounds in cometary ices may also 
have played a role in early mem- 
brane production. Formation of 
membranes is considered a critical 
step in the origin of life, for they are 
required to isolate and protect the 
interior workings of a cell from the 
surrounding medium. Within the 
confines of a membrane, conditions 
such as acidity can be moderated 
and held at a different value from 
that in the surrounding medium, 
nutrients can be concentrated, and 
so on. Although it is uncertain where 
membrane formation falls in the 
sequence of events leading up to the 
origin of life, it is considered an 
absolutely crucial step. When the 
most complex organic molecules 
produced by irradiation of the 
simple mixed molecular ices are 
placed in water, w^iter-insolul)le 
droplets form naturally. This self- 
organizing, lipid-like behavior is 
similar to that found for the organic 
components of the Murchison 
meteorite. The figure shows micro- 
graphs of these nonpolar droplets. 
Both photographs are of the same 
field of view, with the upper photo- 
graph taken in normal light (yellow- 
ish cast) and the lower in fluorescent 



V K (> K K S S IN A S T \i (I I' II V S [ C S 



93 




Fig. h Photomicrographs of the water-insoluble droplets formed when a few 
drops of water are placed on the most complex molecules produced by 
irradiation of an interstellar and cometary ice analog. Both photographs 
show the same field of view, with the upper photograph taken in natural 
lighting and the lower in fluorescence stimulated by ultraviolet. Taken 
together, these photographs show that the self-organizing droplets form 
naturally from these materials and that they also trap luminescent materials 
in their structures. 



emission stimulated by a black light 
(ultraviolet light). The upper photo- 
graph shows the presence of two 
distinct phases, liquid water and 
insoluble droplets. Several of the 
self-organized droplets show intrigu- 
ing structures suggestive of cell 
walls. In comparing the upper and 
lower photographs, it is interesting to 
note that all of the larger and many 
of the smaller droplets have trapped 
blue photoluminescent material also 
produced by the original ice irradia- 
tion. The ability to trap energy 
receptors within these structures is 
also considered a critical step in the 
origin of life, for it provides the 
means to "power" the protocell, as 
photosynthesis does today in plants. 

Point of Contact: L. Allamandola 
(650) 604-6890 
allamandola@ssa1 .arc.nasa.gov 



Spectrum Synthesis of Hot 
Water in Sunspots and 
Selected Cool Stars 

Duane F. Carbon, David Goorvitch 

The objective of this research is 
to define the energy distribution of 
late-type stars so that their physical 
state is better known. Better informa- 
tion about their physical state will 
then lead to improved abundance 
determinations and energy stan- 
dards. Very recently, Partridge and 
Schwenke of Ames Research Center 
completed an elaborate theoretical 
computation of the potential-energy 
surface and dipole moment function 
for H^O. They have used their 
results to predict the positions and 



94 



P K {; K I- S S [ N A S T K V fi V S I t; s 



Space S c i e \ c e E \ t e r p r i s e 



strengths of nearly 308 million 
^H^H'^^O lines. This line tabulation 
is the most complete now available. 
It extends to sufficiently high excita- 
tions so that the spectra of M-stars 
(cool red stars) may be modeled 
with greater accuracy than ever 
l)efore provided the predicted line 
parameters of Partridge and 
Schwenke are themselves accurate. 
Synthetic sunspot spectra have 
been computed using the Partridge 
and Schwenke H2O line list and the 
sunspot umbra! models. In order to 
test the new line list, a careful 
comparison of these synthetic 
spectra and of published high- 
resolution sunspot atlases has been 
carried out. This comparison shows 
that the new H7O line list success- 
fully predicts the sunspot H2O 
spectrum and that it can specify 
where further improvements in the 
line tabulation are necessary. In 
addition, as part of this study, it is 
demonstrated that sunspot atmo- 
spheres are much cooler in their 
shallow layers than predicted by the 
models and that the most recent line 
tabulation for the OH molecule is 
seriously in error. 

Using the new H^O tabulation, 
the extent to which hot stellar 
water blankets the H (1 .65 microns), 
K (2.20 microns), and L 
(3.45 microns) passbands for 
selected K- and M-star model 
atmospheres and, thus, the energy 
distribution of these stars, is also 
investigated. 

Point of Contact: D. Carbon 

(650)604-4413 

clcarbon@nas,nasa.gov 



A High-Altitude Site 
Survey for SOFIA 

Michael R. Haas, Leonhard Pfister 

The Stratospheric Observatory 
for Infrared Astronomy (SOFIA) will 
have a 2.5-meter telescope mounted 
in the aft section of a Boeing 747 SP 
for use in making astronomical 
observations at wavelengths 
between 0.3 and 1600 microns. 
This study investigates the meteoro- 
logical conditions above 41,000 feet 
and finds that there are significant 
weather patterns at these altitudes 
which can adversely affect airborne 
astronomy. 

Experience has shown that high- 
altitude cloud cover and water-vapor 
overburden are the most important 
meteorological factors affecting the 
quality of astronomical observations 
at typical flight altitudes. To deter- 
mine acceptable home base sites for 
SOFIA, these factors must be mea- 
sured over broad geographic 
regions, since a typical flightpath 
covers hundreds of kilometers. The 
only source of such data is satellites 
which provide global coverage of 
Earth over an extended period of 
time. 

This study gathers the best 
available satellite data, shows that 
they corroborate airborne experi- 
ence, and demonstrates that they are 
consistent with well-known global 
circulation, convection patterns, 
and upper-tropospheric/lower- 
stratospheric dynamics. Part (a) of 
the figure (see Color Plate 19 in the 
Appendix) shows the total frequency 
of cloud occurrence as measured by 
the Stratospheric Aerosol and Gas 
Experiment (SAGE II). This instru- 
ment uses a limb occupation 



technique, so it has good vertical 
resolution, but relatively poor 
horizontal resolution. The cloud- 
frequency data have been vertically 
integrated above 41 ,000 feet and 
averaged over the three summer 
months (June, July, and August) for a 
6-year period. Part (b) of the figure 
shows the total frequency of cloud 
occurrence for summer as measured 
by the High-Resolution Infrared 
Radiation Sounder (HIRS) sensors on 
polar orbiting National Oceanic and 
Atmospheric Administration satel- 
lites since June, 1 989. Since this 
satellite is nadir-looking, this dataset 
has much better horizontal resolu- 
tion, but a significantly higher 
minimum optical depth and poorer 
vertical resolution. 

Water vapor is the principal 
molecular absorber throughout 
much of the infrared. Its strong 
vertical stratification at altitudes 
below the tropopause is one of the 
primary motivations for conducting 
airborne observations from the 
stratosphere, where the water-vapor 
overburden is relatively small and 
constant. Furthermore, the negative 
temperature gradients just below 
the tropopause allow rapid vertical 
transport of moisture. When com- 
bined with low, upper tropospheric 
temperatures, this results in the 
formation of extended, thin sheets 
of cirrus clouds. Part (c) shows the 
average tropopause height for the 
summer season, and part (d) shows 
the water-vapor overburdens in 
precipjtable microns as inferred from 
the Microwave Limb Sounder (MLS) 
for the summers of 1992 and 1993. 
Although the positive temperature 
gradient above the tropopause 
strongly increases the vertical 
stability of the atmosphere and 



R (1 R !•: s s 



A S T \( i' 



95 



places a lid on the rapid vertical 
transport of moisture in Earth's 
atnnosphere, there are clearly large 
variations from this simple model. 
The largest variations in high- 
altitude weather over the United 
States occur in the summer because 
the tropopause is near 41 ,000 feet. 
Moreover, different portions of the 
United States lie in different meteo- 
rological regimes in terms of large- 
scale circulation and the potential 
for convection. The eastern Pacific is 
dominated by the downward motion 
associated with the mid-Pacific 
upper-level trough, and by generally 
lower sea surface temperatures. East 
of a line extending from northern 
Mexico along the Rockies into 
Canada, there is upward motion 
associated with the continental 
upper-level ridge. Warm moist air 
from the Gulf of Mexico, and to a 
lesser extent from the Gulf of 
California, promotes convection east 
of the Rockies and in northwestern 
Mexico. Moffett Field, the site 
selected for SOFIA operations, is 
well within the regime of downward 
motion which results in lower upper- 
level cloudiness and lower water- 
vapor overburdens. 

Point of Contact: M. Haas 
(650)604-5511 

mhaas@maiLarc.nasa.gov 



Kepler Mission 
Educational and Public 
Outreach Software 

David Koch 

In response to the strategic goals 
of finding other habitable planets 



and conveying this objective to 
students and the public, an educa- 
tional software package has been 
developed and distributed publicly. 

Experience with educators has 
shown a demand not just for infor- 
mation, but also for science data that 
students could analyze. This has 
been a difficult challenge, since in 
general the latest data from airborne 
and space-borne experiments almost 
always requires sophisticated data 
processing and an ''educated eye'' 
to glean the subtle effects being 
observed. However, the observations 
to be performed by the Kepler 
Mission, which has been proposed 
to the Discovery Program, to detect 
Earth-sized planets are based on a 
straightforward principle that can be 
easily understood even by upper 
elementary students. An interactive 
software package has been devel- 
oped that allows the students to both 
perform simulated observations as if 
they were the astronomers and then 
to analyze the data to determine the 
habitable properties of the planet. 

The Kepler Mission has been 
proposed as a Discovery Mission to 
search for habitable planets. The 
purpose of the mission is to continu- 
ously and simultaneously observe 
1 60,000 stars for evidence of 
planetary transits. When a planet 
passes in front of its star it causes the 
star to dim. When a second transit of 
the same duration and with the same 
brightness change occurs, the exact 
orbital period can be predicted. 
When additional transits are seen, 
one has a high confidence that a 
planet has been detected. From the 
period, the orbital distance of the 
planet from its star can be calculated 
and the characteristic temperature 
of the planet estimated. From the 



change in brightness the planet sizt^ 
can be calculated, and the mass and 
surface gravity can be estimated. 

The software package has 
several components, including the 
following: 

1 . A description of the latest results 
in the search for planets beyond our 
own solar system; 

2. A description of a habitable 
planet, one where liquid water may 
be found on the surface; 

3. A description of the various 
methods, including the capabilities 
and limitations of the different 
detection techniques; 

4. A description of the Kepler 
Mission, the only currently practical 
method for finding Earth-sized 
planets; 

5. An interactive Kepler Mission 
simulator; 

6. A biography of Johannes Kepler, 
a description of his three laws of 
planetary motion, and information 
on his life and times. 

The software is written in 
Hyperstudio^^^ for the Macintosh, 
and the HyperStudio' ^' player is 
included. The software can be 
downloaded from the Kepler Mission 
web site {WWW.kepler.arc.nasa.gov). 
When compressed, the entire 
software package with the player is 
1 megabyte and will fit on a single 
high-density disk; thus it is easy for 
teachers to copy it and carry it into 
the classroom. Use of the software 
does not require an Internet 
connection. 

Point of Contact: D. Koch 
(650) 604-6548 
dkoch@mail.arc.nasa.gov 



96 



P a c K !■: s s IN As r \< o v m v s i c s 



Space S c i t s c e E s t e r p r } s e 



Mid-Infrared Studies of 
Diffuse Interstellar 
Material 

Thomas L. Roellig 

Infrared observations of the 
emission from material located in 
the diffuse regions of space between 
the stars can provide important 
information on its composition and 
energy state. This information is 
crucial to our understanding of star 
formation and galactic evolution. 
Unfortunately, such observations are 
very difficult, largely because this 
emission is very weak and is easily 
drowned out by infrared radiation 
emitted from the telescope optics. 
Furthermore, many of the most 
interesting regions in the infrared 
spectrum are obscured by absorption 
in Earth's atmosphere. 

As a result, a telescope in space 
cooled to temperatures only a few 
degrees above absolute zero is 
needed for these investigations. Such 
a telescope, optimized for infrared 
studies of diffuse regions, was flown 
in 1995 in a NASA/Japanese Space 
Agency collaboration— the Infrared 
Telescope in Space (IRTS) infrared 
survey mission. The Mid-Infrared 
Spectrometer (MIRS), an instrument 
developed and operated jointly by 
NASA Ames and the University of 
Tokyo, was attached to the IRTS and 
operated over wavelengths ranging 
from 4.6 to 1 1 .7 microns. This is an 
idea! spectral region for investiga- 
tions of infrared emission from solid 
material, since it covers many of the 
diagnostic stretching and bending 
modes of candidate interstellar 
material. The IRTS/MIRS telescope/ 
instrument combination achieved 



sensitivities that were orders of 
magnitude better than any instru- 
mentation previously used for diffuse 
mid-infrared emission studies. 

The first data from the MIRS 
have now been analyzed and have 
yielded interesting new information 
about the interstellar material. The 
observed mid-infrared interstellar 
emission was found to be concen- 
trated almost solely in a few emis- 
sion bands (see the figure). This was 



true both for those regions located 
within the plane of the galaxy and 
for those regions above the galactic 
plane. The remarkable similarity of 
the spectra from the different regions 
indicates that there is little chemical 
differentiation between these 
regions. The spectral features 
themselves are indicative of emis- 
sion from a class of hydrocarbons — 
polycyclic aromatic hydrocarbons 
(PAHs). In the scenario envisaged to 



xlO^Wm ^sr ^ 



14 



12 - 



10 



(A 



0) 

u 

CO 

r 

CO 



4 - 




MM 



"I'rr I r < It I I I I 1 
6 7 



I I I I 



Ttih-1-I -i^-i-iM 



8 9 

Wavelength (|im) 



10 



11 



12 



Fig. /. Spectra obtained from the Mid^lnfrared Spectrometer on the Infrared 
Telescope in Space mission. For all spectra, the galactic longitude was 
approximately 50''. The different lines correspond to different galactic 
latitudes, b. The solid line indicates emission from -O" 40' < b < rr, the 
dashed line emission from r<b <2^, the dotted line emission from 
2° < b < 3'', and the dot-dashed line emission from .^" < b < S"". 



P K (I (; K K S S IN As T \< P IM S 1 C S 



97 



operate in the interstellar medium, 
these molecules absorb ultraviolet 
photons and then emit this energy in 
the infrared. The spectra obtained by 
the MIRS show that the mid-infrared 
emission from the diffuse regions in 
the galaxy can be attributed almost 
solely to this material. The MIRS 
data also show that the previously 
unexplained excess emission 
observed in the mid-infrared by the 
earlier Infrared Astronomical Satel- 
lite (IRAS) mission can be attributed 
to these emission bands. 

Interestingly enough, a compari- 
son of the spatial distribution of 
these mid-infrared emission features 
with the emission at the much 
longer wavelength of 1 00 microns 
observed with the Infrared Astro- 
nomical Satellite mission indicates a 
very close correlation. Since the 
100-micron emission is not expected 
to arise from PAHs, but instead from 
much larger grains of solid material, 
this tight spatial correlation must 
mean that these two different kinds 
of interstellar materials are mixed in 
a constant ratio throughout the 
different interstellar regions in 
the galaxy. 

Point of Contact: T. Roellig 
(650) 604-6426 
troellig@maiLarc.nasa.gov 



Infrared Observations of 
GOJ 8-0.04 

Janet P. Simpson, Sean W. j. Colgan, 
Angela S. Cotera, Edwin F. Erickson, 
Michael R, Haas, Mark Morris, 
Robert H. Rubin 

Radio images of the center of 
our Galaxy show a number of 
features that are unique in the 
Galaxy and that are very difficult to 
explain. Some of these are demon- 
strated in the figure, which is a 
20-cm Very Large Array radio image 
by F. Yusef-Zadeh and M. Morris, 
overlaid with contours showing the 
locations of the Galactic Center 
molecular clouds. A quarter of a 
degree from the Galactic Center 



there is a long thin ''Arc/' also called 
the Straight Filaments, that crosses 
the Galactic Plane at right angles. 
The length and thinness of the 
synchrotron radiation emitting Arc 
(and other arcs not in the figure), 
suggest that the Galactic Center 
region has a strong magnetic field; 
however, the excitation and source 
of the relativistic electrons emitting 
the synchrotron radiation is not 
known. 

The Galactic Center H II region, 
GO. 18-0.04, also called the ^'Sickle," 
is located where the Arc crosses the 
Galactic plane. The Sickle appears 
to be the ionized edge of the dense 
molecular cloud whose contours are 
shown in the figure. In the past, the 
source of ionization of the Sickle has 



-28 40' - 



Galactic plane 



^ Arched 
V filaments 

W9^ 



o 

o 




-2roo' - 



45'^30^ 1 7^43'"0® 42'"30^ 42"^0® 

Right ascension 



Fig. L The 20-cm Very Large Array radio image, overlaid with contours 
showing the locations of the Galactic Center molecular clouds. Sgr A is the 
Galactic Center GO. 18-0,04 (the Sickle) and the Pistol are labeled. 



98 



P K (i R !•: S S [ N As T k F H V S I C S 



Space S c i e s c e E s t e r f r i s ^ 



been ascribed to both the interaction 
of the cloud with the magnetic field 
of the Arc and to the hot stars in the 
adjacent star cluster, AFGL 2004, 
also known as the ''Quintuplet 
Cluster." There is a smaller region of 
thermal gas known as the "Pistol"; 
the star cluster AFGL 2004 lies 
between the Sickle and the Pistol. 
Because of the great obscuration 
toward the Galactic Center, it is not 
possible to see these regions at 
w^avelengths visible to the eye; 
instead, one must use infrared and 
radio telescopes to view the region. 
The objective of this study is to 
ascertain the relative locations of the 
stars, the ionized and molecular gas, 
and the sources of gas excitation and 
dust heating. 

Far-infrared lines of C+, N+, 
N++, GO, ()++, Si+-i-, and S-\-+ were 
measured with the Cryogenic 
Grating Spectrometer on the Kuiper 
Airborne Observatory. For the hi II 
region, electron densities for both 
the 0++ and S++ zones were 
estimated, and abundance ratios 
were calculated for C)-i-+/S++, N++/ 
0++, N+/N-f+, N+/0++. Si+/S++, 
and S+-i-/hH+. For the less excited gas 
in the interface region between the 
\-\ II region and the molecular cloud, 
the incident photon flux and densi- 
ties were estimated from the C-i- 
and C)0 lines and the far-infrared 
continuum. 

Near-infrared images of the 
region were taken by A. Cotera as 
part of her Stanford University Ph.D. 
thesis (1995), supported by a NASA 
Graduate Student Research Fellow- 
ship at Ames. From the 1 .65 and 
2.1 micrometer photometry of stars 
in the field and from the FHydrogen 
Br gamma images, the extinction to 
the region could be estimated. The 



extinction is fairly uniform over the 
region and is consistent with a 
location in the Galactic Center. 
There is no indication of the dense 
molecular cloud that had previously 
been suggested by other researchers 
to lie between Earth and the region 
in the images of the star field. The 
previous suggestion of excitation of 
the Sic kle by interaction of the 
molecular cloud with the magnetic 
field recjuired that the cloud, which 
recedes at 25 kilometers per second 
away from Earth, f)roduce the 
observed 20- to 1 20-kilometers-fx^r- 
second ionized gas in the FH II region 
by running into the magnetic flux 
tubes of the Arc. Thus, both the Arc 
and the ionized gas would l^e on the 
far side of the cloud from Earth in 
this scenario. Since the near-infrared 
measurements of the stars and the 
ionized gas demonstrate that the gas 
must be on the near side of the 
cloud, this early suggestion cannot 
be correct. 

The observed lines were com- 
pared with predictions of models of 
slab H II regions ionized by a cluster 
of luminous stars. This comparison 
indicates that the sourc e of ioniza- 
tion of the Sickle is probably the hot 
stars of the AFGL 2004 cluster, but 
the Pistol has a lower ionization than 
its proximity to the cluster would 
suggest. More likely, the Pistol is 
ionized by the extremely luminous 
star that the Ames researchers found 
to be located near the center of its 
arc. Since the positions at some 
distance from AFGL 2004 have 
higher ionization than is predicted 
l)y the models, it is likely that there 
are additional hot stars scattered 
throughout the region. The low 
excitation lines and far-infrared 
continuum from the molecular cloud 



and interface region between the 
cloud and H II region are also 
consistent with heating and excita- 
tion by the same stars. The electrons 
in the Arc probably originate in the 
ionized gas of the H II region and 
are excited to relativistic energies 
by magnetic reconnect ion of the 
magnetic field lines of the Arc and 
the molecular c loud. 

Point of Contact: J, Simpson 

(650)604-1613 

simpson@cygnus.arc.nasa.gov 



New 3.405-Micron 
Interstellar Emission from 
Organic Hydrocarbons 

Gregory C. Sloan, Jesse D. Bregman 

Polycyclic aromatic hydrocar- 
bons (PAHs) are a form of organic 
hydrocarbon identified by their 
complex spectral emission in a rich 
variety of astronomical sources. This 
spectrum includes several emission 
features between 3 and 1.3 microns, 
originally described as the unidenti- 
fied infrared emission features. Louis 
Allamandola and his Ames collabo- 
rators identified the carriers of this 
spectrum by their laboratory analysis 
of PAHs. These molecules typically 
consist of 20-50 carbon atoms 
arranged in a flat matrix of aromatic 
rings, with hydrogen atoms bonded 
to the carbon on the edge. The 
C-H bonds produce most of the 
spectral features observed from 
the ground; stretching modes 
produce features at 3.3 microns and 
nearby; bending modes produce 
features at 8.6 microns and in the 



P K <i \< !■: S S [ \ As T R O 1' [| V S I c s 



99 



1 1 .2-1 2.7-micron region. These 
organic molecules nnay represent 
more than 1 57(> of all interstellar 
carbon, so a better understanding of 
their composition would give us 
insight into the evolution of interstel- 
lar carbon material. Because PAH 
spectra are seen in so many sources, 
a better understanding of the emis- 
sion mechanisms would make them 
very powerful probes of the interstel- 
lar medium. 

PAH spectra are observed 
typically at the edge of ionized 
regions. One such region, the Orion 
Bar, is a transition region between 
the Orion Nebula, ionized by the 
ultraviolet energy from the Trape- 
zium, and the molecular cloud to 
the southeast. These transition 
regions have strong gradients in 
ultraviolet flux decreasing from the 
ionized zone to the molecular 
region. The Bar presents the observer 
on Earth with an edge-on geometry 
which makes it an excellent labora- 
tory to study how PAH emission 
varies with excitation conditions. 

The 3-micron PAH emission in 
the Orion Bar was observed using a 
long-slit spectrometer at the United 
Kingdom Infrared Telescope in 
October 1995. These spectra allow 
one to trace the strength of the 
3.29-micron PAH feature, which 
dominates the spectrum and arises 
from a ground-state aromatic C-H 
stretch; the weaker features observed 
at 3.40 and 3.51 microns; and the 
underlying emission plateau which 
stretches from 3.3 to 3.6 microns. 
These features peak on the ioniza- 
tion front, and drop in strength into 
both the ionized and molecular 
regions. The features decrease into 
the ionized region because the harsh 



ultraviolet field is destroying the 
PAHs there. They decrease into the 
molecular region because the 
ultraviolet flux which excites the 
spectral features is being attenuated 
by the high optical depths of the 
neutral and molecular gas in the Bar. 
in general, these features decrease 
into the molecular region exponen- 
tially, with a 1/e scale height of 
approximately 10 arcseconds. 
However, the PAhH plateau, and the 
3.40- and 3.51 -micron features show 
a layer of enhanced emission 
roughly 10 arcseconds behind the 
ionization front, right where emis- 
sion from H2 also peaks. This 
enhancement is especially notice- 
able in the 3.40-micron emission 
feature. 

A careful analysis of both the 
shape and the strength of the 
3.40-micron feature reveals that it 
actually consists of two spectrally 
and spatially distinct components. 
The main component peaks at 
3.395 microns and behaves spatially 
very similarly to the PAH plateau 
and the 3.51 -micron feature. The 
secondary component peaks at 
3.405 microns, and its emission is 
concentrated 10 arcseconds behind 
the ionization front, again, right 
where the H2 emission peaks. 

The spectral properties can be 
interpreted in terms of recent 
laboratory analyses of PAHs at 
NASA Ames by Bernstein, Sandford, 
and Allamandola and by Joblin, 
Tielens, Allamandola, and Geballe. 
Joblin et al. analyzed PAHs with 
attached methyl sidegroups, which 
contain aliphatic C-H bonds and 
produce an emission feature close 
to the position of the secondary 
component, 3.405 microns. 



Bernstein et al. analyzed what they 
describe as H-PAHs, or PAHs with 
additional H atoms attached to the 
carbon on the periphery of the 
PAHs. These atoms convert aromatic 
C-H bonds to aliphatic C-H bonds, 
producing an emission feature at 
3.395 microns, as well as the PAH 
plateau and emission feature at 
3.51 microns. In fact, their labora- 
tory spectrum of hexahydropyrene 
provides the closest match yet 
attained to the astronomical PAH 
spectrum. 

The evidence is growing that the 
3.40-micron feature arises from an 
aliphatic C-H bond. Both of the 
components at 3.40 microns that 
we have identified arise from such a 
bond, although the bond resides in 
different carriers. Based on the 
differences in spatial behavior 
between these components, it 
appears that the methyl sidegroups 
are easily destroyed beyond the 
molecular hydrogen emission layer, 
whereas the H-PAHs survive right up 
to the ionization front. 

Point of Contact: J. Bregman 

(650)604-6136 

jbregman@mail.arc.nasa.gov 



100 



P R {) C, K K S S ! N As T \< I* 1! Y S IC S 



S P A ( F S { ! i AM E \ T E R P R I S E 



Assessment of the Cassini 
Command and Data 
Subsystem 

Edward A. Addy 

Formal methods have been 
promoted as an ettective means of 
developing software that requires a 
high degree of assurance. Yet to 
dale, formal methods have not 
significantly affected software 
development practices. The difficulty 
of maintaining the fidelity of the 
formal model with the system as the 
system undergoes changes in 
requirements and specifications 
during development has contributed 
to its failure to influence the devel- 
opmental practices. The goal of this 
project is to determine if the use of 



Formal model 

Model checking 
Theorem proving 



Test cases 




System 



Test data 



Requirements 
Specifications 
Implementation 




F/^^ /. Use of theorum-provin^ in 
generating test cases to maintain 
formal model fidelity during system 
development. 



theorem-proving within a formal 
model can be used to generate test 
cases for testing the actual systen^j 
being modeled. The union of formal 
models with test-case generation will 
help to alleviate the difficulty of 
maintaining fidelity, and will provide 
useful information in the form of test 
cases to the development team. 

The functioning of the scheduler 
and interrupt handler through 
analysis of the Cassini Command 
and Data Subsystem (CDS) source 
code has been used as a case study. 
A formal model of this part of the 
CDS software was created, using the 
formal specification language, 
Prototype Verification System (PVS), 
which is based on a classic, typed 
higher-order logic. The PVS system 
includes support tools and a theorem 
prover. This follows on from previ- 
ous research in which formal model 
checking was used to generate test 
cases for the actual system. The test 
data and the model checking have 
been used jointly to maintain the 
fidelity of the model with the actual 
system as it is being developed. This 
project has used theorem-f^oving 
within the formal model in a similar 
fashion, to generate test cases for the 
CDS and to increase the fidelity of 
the model to the CDS, as illustrated 
in the figure. 

The source code from the 
Cassini CDS related to the scheduler 
and interrupt handler functions was 
analyzed. The formal modeling 
process identified that the millisec- 
ond interrupt was not functioning 
properly. The development team 
modified the scheduler, improving 
its performance and eliminating the 
potential for improper memory 



allocation. A prelin^inary PVS model 
of the scheduler and interrupt 
handler functions has also been 
developed. 

Point of Contact: E. Addy 
(304) 367-8353 
edward.addy@ivv.nasa.gov 



Automatic Telescope 
Project 

)ohn Bresina 

The primary ol)jective of the 
Automatic Telesc ope Project is to 
provide advanced scheduling and 
automation infrastructure so that the 
benefits and performance of single- 
user automatic telescopes can be 
made available to a large commu- 
nity of users, with no additional 
people in the loop. In addition. 
Web-based assistance is provided 
for "service observing/' which is 
becoming the dominant operations 
mode for large-scale, sophisticated 
telescopes. The benefits are a 
dramatic reduction in telescope 
operating costs and an increase in 
l)oth scientific productivity and ease- 
of-use. The Associate Principal 
AstrontMi^er (APA) enables electronic 
recjuest sul^mission, automatic 
loading and scheduling, and elec- 
tronic data return, all under the 
control of a single astronomer (see 
the figure for an illustration of the 
architecture). The advanced loading 
and scheduling technic^ues provide 
fair allocation of telescope tinu^ and 
support high-quality data collection. 

The APA software tools have 
been extensively tested by G. Henry 
at Tennessee State University (TSU) 



1» K E) (: k !■ N S I \ S r A c i: 



101 




Astronomers 



Fig. 1 . Associate Principal Astronomer architecture. 



and by the Fairborn Observatory 
staff. The scheduler has been 
executing nightly since January 1 996 
for Fairborn telescope 18. The loader 
has been undergoing nightly testing 
for the same telescope since April 
1 996, and the Service Observing 
Associate (SOA) is undergoing 



testing on the UK Infrared Telescope 
(UKIRT). There was enthusiastic 
reception of the APA from UKIRT, 
the Hands-On-Universe project, the 
Four College Telescope, and UK's 
New^ Generation Astronomical 
Telescopes Project. John Davies 
carried out a successful "live test" 



of the SOA at UKIRT and demon- 
strated the system to senior staff of 
the multinational Gemini project 
(and others). 

Point of Contact: J. Bresina 
(650) 604-3365 
bresina@ptolemy.arc.nasa.gov 



102 



P H (i R !•: S S IN S P A C K T 



}■: c 



N I, (i i [•; s 



Space S c i e \ c e E \ t e r p r i s e 



Guide Star Tracker for 
Gravity Probe B Relativity 
Mission 

John H. Goebel 

The ^uide star tracker for the 
Gravity Probe B (GPB) Relativity 
Mission is used to define the inertia! 
reference frame against which the 
relativistic effects of a satellite in a 
polar Earth orbit can be measured. 
The tracking telescope is cooled to 
superfluid helium temperature 



(1 .87 kelvin) in order to accommo- 
date the gyroscope readout detectors 
of the Superconducting Quantum 
Interference Device (SQUID). 
Because both the gyroscope and 
guide telescope are rigidly mounted 
together in the liquid-helium-cooled 
portion of the satellite, the starlight 
sensing detectors and preamplifier 
electronics must function without 
disturbing either the gyroscopes or 
the telescope wavefront. Tracking 
accuracy of the order of 10 milliarc- 
seconds is required, implying that 




Fit;. 1. Revision B Hybrid Circuit. This circuit is OAOO inches in dianwter 
and includes two circular silicon photodiodes and four square silicon field 
effect transistors. Light from the guide star is focused on the detectors which 
convert the light to electrical current which in turn is amplified by the 
transistors. The circuit is thermally isolated from the cryogenic environment 
so that it can operate at an elevated temperature of 70 kelvin. 



thermal disturbances must be kept to 
a minimum. 

The Space Technology Branch 
has contributed to the star sensor 
development effort for GPB by 
assisting in the design, construction, 
development and testing of the 
photodiode and cooled electronics. 
This effort is a collaboration with 
scientists and engineers at Stanford 
University, Marshall Space Flight 
Center, Goddard Space Flight 
Center, Ames Research Center, 
Lockheed-Martin Corporation, and 
various suppliers of component 
parts. 

A photograph of the Revision B 
hybrid electronics circuit for the star 
sensor that is mounted in the 
cryogenic vacuum vessel is shown 
in the figure. The sensor has been 
demonstrated to meet the require- 
ments for functioning in the cryo- 
genic environment by operating at a 
somewhat elevated temperature of 
70 kelvin with a power dissipation 
not exceeding 2 milliwatts. Opera- 
tion of the sensor at temperatures 
down to 30 kelvin has been demon- 
strated at power dissipation less than 
1 milliwatt. 

Point of Contact: J. Goebel 

(650)604-3188 

jgoebel@mail.arc.nasa.gov 



Pulse-Tube Cryocooler 
Development 

Peter Kittel 

Pulse-tube coolers have unique 
advantages over the other coolers 
that are usually considered for NASA 
space missions. Because they have 



P K {; K [•; s s 1 \ S p A c k T !•; 



103 



no moving parts in the cold part of 
the cooler, they have inherently very 
long lifetimes and very low levels of 
vibration. However, they are a 
relatively new technology and 
require further development to 
achieve the thermal efficiency of 
more mature coolers such as Stirling 
coolers. The first figure shows the 
main components of a pulse-tube 
cryocooler. 

The compressor (at room 
temperature) provides an oscillating 
gas pressure which moves gas into 
and out of the cooler. The regenera- 
tor consists of a tightly packed bed 
of fine screens that absorb any 
temperature oscillations in the gas. 
The compressor end of the regenera- 
tor is at room temperature and the 
other end is cold. The pulse tube 
isolates the cold junction from the 
orifice and reservoir at room tem- 
perature. A properly adjusted orifice 
creates a special relationship 
between the pressure oscillations in 
the pulse tube and the gas flow such 
that heat is drawn from the cold 
junction and deposited at the orifice. 

jn FY96 great progress was 
made in understanding and improv- 
ing the operation of two of the main 
components of the cooler — the 
regenerator and the orifice. 

For the screen mesh in the 
regenerator to adequately absorb 
temperature oscillations in the gas 
passing through it, it must have a 
large surface area and a large mass. 
These requirements suggest a large 
number of very fine screens and a 
high impedance to the gas flow. The 
regenerator must not be too restric- 
tive, because this would reduce the 
pressure and flow available for 
cooling at the pulse tube. It is crucial 
to find the best compromise between 




Compressor 



Regenerator j 

Coid junction 



Pulse tube 




Reservoir 



Fig. 1. Components of j pulse tubc-cryocooler. 



absorbing temperature oscillations 
and allowing adequate flow. 

The behavior of this region of 
the cooler has been difficult to 
predict; measurements of pressure 
drop with oscillating flow have 
disagreed with predictions based 
on steady-flow data. Therefore, 
measurements were made to charac- 
terize the screen mesh of the regen- 
erator under oscillating flow 



conditions that were similar to those 
that would exist in actual use. These 
measurements showed that the 
pressure drop at frequencies near 
60 cycles per second was 20-30'X) 
less than that predicted from steady- 
flow data. These improved data 
(shown in the second figure) will 
allow much better optimization of 
the regenerator screens. 




100 



Reynold's number 



Fig. 2. Pressure drop in a 325-mesh regenerator. Solid curve is steady-flow 
delta. The data points are the instantaneous pressure drop during 60-cycle 
runs for a variety of amplitudes. 



104 



P K (; R [■: s s 



[ N S r A c I- 



H C M N L C [ I- S 



Space S c i e s ( e E s t e r p r i s e 



The orifice serves to adjust the 
relationship between oscillating gas 
pressure and gas flow in the pulse- 
tul)e section. An analysis of the 
system shows that much more 
cooling power could be obtained if 
sufficient adjustment of this relation- 
ship could be achieved. However, 
the range of adjustment available 
from a simple orifice is not enough 
to allow the utilization of the extra 
cooling power that the system is 
capable of. What is needed is a 
device that has ''inertance" in 
addition to the dissipation of the 
orifice. To use an electrical analogy, 
the ''resistance'' represented l)y the 
orifice needs to be replaced by a 
device with "inductance/' A device 
with the inertance needed turns out 
to be a long narrow tube; typically it 
would be 2-:] millimeters \n diam- 
eter and several meters long. 

Preliminary testing has been 
performed with such an inertance 
tube in place of the orifice, and the 
cooler performed considerably better 
than it did with the orifice; this 
shows that the analysis is on the 
right track. More improvement is 
expected as understanding of the 
operation of the inertance tube is 
gained and its parameters are 
optimized. 

Point of Contact: P. Kittel 
(650) 604-4297 
pkittel@mail.arc.nasa.gov 



Automated Space System 
Experimental Testbed 
Project 

Christopher Kitts 

The Automated Space System 
Experimental Testbed (ASSET) 
project is an Internet-based, justified- 
operations control system for small 
satellites. The project's objective 
was to establish a global spacecraft 
control network in order to (1) pro- 
vide operational support for a variety 
of low-cost missions, and (2) provide 
a high-risk, low-inertia testbed for 
validating spacecraft autonomy 
strategies. 

The operational system provides 
an automated link between a variety 
of users (including principal investi- 
gators, students, operators, and the 
public) and a multitude of spacecraft 
payloads. The ASSET system is 
capable of accepting goal-level 
experimental parameters and 
planning and scheduling system and 
spacec raft activities, and of conduct- 
ing mission and health-related 
tracking, telemetry, commanding 
operations with the spacecraft, and 
delivering the experimental data to 
the original experimenter. To do this, 
the ASSET system components 
include a central mission control 
center, geographically distributed 
ground stations, a variety of space- 
craft, Internet and HAM radio 
communications links, and numer- 
ous payload components. 

ASSET provided an unparalleled 
opportunity for rapid experimenta- 
tion with revolutionary autonomy 
technologies in a real-wodd, com- 
plex space system. The initial ASSET 
implementation is supporting the 



New Millennium Program 
Autonomy Technical Integration 
Experiments concerning data 
distribution, principal investigator 
interfacing, beacon operations, 
ground system automation, and fault 
management. 

Point of Contact: C. Kitts 

(650)604-0149 

kitts@ptolemy,arc.nasa.gov 



Amphion and 
Meta-Amphion 

Michael Lowry 



Most of today's programming 
problems involve a programmer 
building special domain-oriented 
software libraries (modules that 
perform a specific function) for every 
domain. Then users must make 
softw^are program calls to these 
modules in order to get the work 
done. The correctness of these 
programs is difficult to prove. 
Amphion automates the use of these 
software libraries. It enables a user 
to state a problem in an abstract, 
graphical visual programming 
notation, rather than requiring the 
user to construct a solution by 
manually composing software 
components. Amphion then auto- 
matically generates a program 
consisting of calls to the library 
components to solve a specific 
problem. Moreover, as part of this 
generation process, which is done 
through automatic theorem-proving, 
Amphion produces a mathematical 
proof that the program is correct. 
Formal mathematical methods are 



1' ^ «* *i f^ !■: s s IN Si' \ c !■: T v. c ii n o i. o {, [ [■: s 



105 




File Edit Options Help 




Fi^. h Amphion-generated science-opportunity visualizer for the 
Ccissini Mission. 



applied for program verification. If 
performed manually, this formal 
proof is costly and only becomes 
cost-effective for safety-critical 
systems. 

Meta-Amphion is a system that 
will enable software-library develop- 
ers to create and maintain their own 
specialized Amphion systems. In 
essence, Meta-Amphion is the next- 
generation appi ication-generator 
technology. Using Amphion, it takes 
on average an order of magnitude 
less time for a user to develop a 
domain-oriented problem specifica- 
tion than to manually generate and 
debug a program. Equally important, 
a user does not need to learn the 



details of the components in a 
software library. This removes a 
significant barrier to the use of 
NASA's software libraries. The figure 
shows a snapshot of a science- 
opportunity visualizer for the Cassini 
Mission. The code for the visualizer 
was generated automatically by 
Amphion. 

The benefit of Meta-Amphion is 
to enable the widespread develop- 
ment of specialized Amphion 
systems, thus realizing the dual 
benefits of automated formal meth- 
ods and domain-specific automatic 
programming. Meta-Amphion is the 
set of tools and infrastructure needed 
to elevate programming from the 



code level to the declarative, 
specification level. The generic 
Amphion system is nearly mature, 
having gone through an extensive 
alpha/beta version that has already 
been tested by customers. 

Point of Contact: M. Lowry 
(650) 604-3369 
lowry@ptoIemy.arc.nasa.gov 



Focal-Plane Sensor Array 
Development for 
Astronomy in Space 

Mark E, McKelvey, 
Robert E. McMurray, Jr., 
Craig R. McCreight 

A team at Ames Research Center 
is evaluating the suitability of a 
number of detector and readout- 
device technologies for space- 
infrared astronomy applications. This 
program aims to foster development 
of focal plane array (FPA) detector 
technology that will allows reliable 
background-limited infrared (IR) 
astronomy from space-based plat- 
forms such as the Space Infrared 
Telescope Facility (SIRTF). Advances 
in microelectronics to provide 
operation of low-noise, low-power 
devices in the radiation environment 
found in Earth orbit are keys to the 
success of this program. 

Evaluations of state-of-the-art 
Impurity Band Conduction (IBC) 
focal plane detector arrays produced 
by several manufacturers are con- 
tinuing for applications in the 
5_30.micron wavelength range. 
These detectors rely on a thin, highly 
doped IR-active layer to provide 



106 



K () (; R K S S 1 N 



S 1' A C H T K C H N L (i i K S 



.V /' I ( f S ( I F\ C f E S T E R P R I S E 




Fig. 1. 256 X 256 element infrared detector drray. 



high quantum efficiency from a 
small detector volume, minimizing 
the ionization cross section for 
cosmic ray events. A high-purity 
blocking layer prevents excessive 
dark current, despite the high doping 
levels in the IR-active layer. IBC 
arrays also exhibit w^ider spectral 
response than alternative photo- 
conductor (PC) architectures, 
without many of the performance 
anomalies associated with PC 
devices. IBC architectures are well 
suited to modern epitaxial fabrica- 
tion methods, and the technology 
has progressed to the point where 
large-format hybrid FPAs that are 



sensitive to IR wavelengths as long 
as 28 microns can be reliably 
produced. 

Work in FY96 was concentrated 
on evaluating the performance of 
large-format (to 256 x 256 detector 
elements; see figure) state-of-the-art 
IBC detector arrays in simulated 
on-orbit radiation environments, 
using the 76-inch Crocker Cyclotron 
on the campus of the University of 
California at Davis to provide a high- 
energy proton beam for device 
testing. These tests, in collaboration 
with astronomers from Cornell 
University and from the University of 
Arizona, have shown a marked 



difference in radiation-damage 
susceptibility between similar arrays 
from different manufacturers. This 
damage is manifested as dose- 
dependent activation of dark current 
that could be of crucial importance 
to the Wide-Field Infrared Explorer 
(WIRE) or SIRTF science missions. 
The results of the Ames tests are 
being fed directly back into the 
manufacturing process in order to 
help advance the state of the art. 

Improved IBC devices from a 
joint Ames/Santa Barbara Researc h 
Center development are presently 
being evaluated. These 256 x 256 
arrays utilize a simfjiifieci unit cell 
architecture and new sampling 
techniques to achieve read noise 
levels as low as 1 2 equivalent 
electrons. These devices are proto- 
types for the flight detectors to be 
used on the SIRTF Infrared Array 
Camera (IRAC). The Ames group will 
continue to play an important role in 
the definition and execution of the 
IRAC detector acceptance and 
qualification test program. 

Other recent efforts involve 
hardware and software upgrades to 
the suite of test instrumentation 
maintained in the detector labora- 
tory. Recent upgrades have allowed 
easier manipulation of the data files 
collected from these large-format 
detector arrays. A project is presently 
under way to enhance the capability 
of the data acquisition system to 
handle arrays as large as 1024 x 
1024 elements. Continuous software 
development has provided improved 
test flexibility, automation, and 
repeatability. Work on maturing IBC 
technology and improved readouts, 
along with other sensor technologies 



I' H (Mr K I- S S IN S I' A C K T !■ C fl \ I. i\ 1 l' S 



107 



in ongoing efforts, will enhance 
overall IR focal-plane performance 
in NASA space- and ground-based 
applications. 

Point of Contact: M. McKelvey 
(650) 604-6643 
mmckelvey@mail.arc.nasa*gov 



Intelligent Execution for 
Autonomous Spacecraft 

Barney Pell 

An executive system is an event- 
driven and goal-oriented agent 
that coordinates run-time activity 
among a set of components involved 
in a control system. As such, it 
provides a language and a frame- 
work in which software designers 
can express how planning, control, 
diagnosis, and reconfiguration 
capabilities are to be integrated into 
an autonomous system. It can 
request and execute plans involving 
concurrent activities among sub- 
systems, in the presence of uncer- 
tainty in the success, timing, or 
outcomes of activities. A smart 
executive provides a language for 
expressing goal-decompositions and 
resource interactions. When inter- 
preting this language at run-time, the 
executive automates the decomposi- 
tion of goals into smaller activities 
which can be executed concur- 
rently, thus automating aspects of 
the labor-intensive sequencing 
function in spacecraft operations. 

This task investigates the 
development of increasingly intelli- 
gent execution systems, their role in 
autonomous control systems, and 
their practical application as an 
enabling technology for autonomous 



space missions. We are currently 
grounding the research by develop- 
ing the smart executive, part of the 
Remote Agent (RA) being developed 
for the Deep Space One (DSl ) 
mission, part of NASA's New 
Millennium Program (NMP). DSl 
will autonomously navigate to 
asteroids and comets and will be 
able to achieve its mission in the 
face of a wide variety of faults, 
including total loss of control from 
Earth. 

As an integrating technology, an 
executive enables each software 
component to be expressed indepen- 
dently, but it controls the interac- 
tions among those components so 
that the overall activity of the system 
achieves global constraints as 
dictated by a high-level plan. This 
dramatic increase in modular 
software design reduces develop- 
ment and integration costs and 
increases software reuse both within 
and across missions. The goal- 
directed task-decomposition auto- 
mates much of the labor-intensive 
sequencing process in spacecraft 
operations. This capability also 
enables a new generation of adap- 
tive controllers, in which different 
control modes are developed that 
are switched by a smart executive 
based on feedback from the 
environment. 

The DSl mission will validate 
the smart executive and the entire 
RA in the context of actual space 
flight. The architecture will serve as 
the baseline for the next generation 
of space missions, and will be 
deployed to a wide variety of 
manned and unmanned spacecraft; 
it will also increase the level of 
autonomy in other control systems, 
such as aircraft and factories. 



FY96 accomplishments include 
completion of the rapid prototyping 
effort in November 1995, demon- 
stration of the prototype, and 
presentation of the smart executive 
to the NMP in the December 1 995 
Technology Readiness Review, 
which accepted the RA architecture 
to be developed for the DSl mission. 

Point of Contact: B. Pell 
(650) 604-3361 
pell@ptolemy.arc.nasa.gov 



New Millennium Program 
Deep Space 1 Flight 
Software Program 
Management 

Scott Sawyer 

The New Millennium Program's 
Deep Space One (NMP DSl ) Flight 
Software team comprises staff from 
Ames and the Jet Propulsion Labora- 
tory (JPL) (which also includes 
individuals from the Air Force 
Phillips Laboratory, TRW, Carnegie- 
Mellon University, and Thinkbank, 
Inc.). The team is developing flight 
software (FSW) for the first mission 
of the New Millennium Program 
(NMP). The Ames team (along with 
part of the JPL team) is dedicated to 
developing and testing the ''Remote 
Agent" architecture of the FSW. The 
remote agent (RA) integrates rule- 
and model-based fault detection, 
isolation, and recovery; advanced 
planning and sequence management 
capabilities; and a system executive 
that serves as an autonomous agent 
controlling and coordinating the 
satellite's behaviors during normal 



108 



P K C R H S S IN S P A C K T K C H fi L G I f: S 



Space S c i t: v o E \ } t n p r i s t: 



and abnormal operations. The RA is 
capable of responding autonomously 
to system and payload faults and 
malfunctions, establishing contact 
with ground operations personnel 
when necessary, and replanning 
portions of the mission onboard to 
accommodate unanticipated 
changes in spacecraft performance 
or objectives. The DS1 mission will 
be the first ever to fly an onboard 
autonomous agent flight system. 

Program management objectives 
are (1 ) managing Ames and JPL team 
personnel; (2) defining development 
schedules, descope options, and 
contingency plans when necessary; 
(3) preparing and presenting status 
reviews and briefings to program 
management, Center management, 
and review boards; (4) making 
budget and workforce projections; 
and (5) acquiring the necessary 
resources to allow the team to 
accomplish its goals. Technical 
objectives include flight/ground 
interface design, RA design over- 
sight, and the performance of 
engineering duties related to 
autonomous systems such as 
defining command and telemetry 
system requirements, fault protection 
system policies, and FSW validation 
policies. 

Successful development of an 
RA FSW architecture for NMP 
DS1 will identify new avenues 
of research which will guide the 
growth and development of the next 
generation of ''thinking" spacecraft. 
The RA architecture will be 
expanded to be used in a wide 
variety of manned and unmanned 
vehicles. By carefully selecting 
applications projects to accompany 



the research, Ames can effectively 
focus the research efforts on practi- 
cal problems at the cutting edge of 
technology. 

Point of Contact: K. Swanson 

(650)604-6016 

kswanson@mail.arc.nasa.gov 



P K (; K K S S IN S f^ A C E: T K C II N I. i\ \ K S 



109 



H I M AS E X P I R A r I \ A A D 
D FY E L P M E S r OF S P A ( E E \ T E R P R I S F 







Overview 

NASA's Human Exploration and 
Development of Space (HEDS) 
enterprise brings the frontier of space 
fully within the sphere of human 
activity. Ames supports the HEDS 
enterprise by conducting research, 
managing spaceflight projects, and 
developing advanced technologies. 
An objective of these efforts is to 
seek knowledge of physicochemical 
and biological phenomena that can 
be fully explored only at very low 
gravity levels. This objective 
embraces the quest for knowledge 
of the role and influence of gravity in 
living systems — one of the elements 
of NASA's astrobiology activities. 
A second objective is to develop 
technologies for advancing human 
exploration of space and achieving 
routine space travel. A complemen- 
tary objective is to apply knowledge 
generated in pursuit of these objec- 
tives, where feasible, to enrich life 
on Earth. During FY96, numerous 
research and technology efforts that 
addressed the following goals of 
the HEDS enterprise were 
accomplished: 

• Increase knowledge of nature's 
processes using the space 
environment. 

• Advance human exploration of 
the solar system. 

• Achieve routine space travel. 

• Enrich life on Earth through 
educational, commercial, and 
technological opportunities in 
space. 

New knowledge and an increase 
in the understanding of nature's 
processes related to the influence 
of gravity on living systems are 



acquired by a two-pronged 
approach: (1) Research is conducted 
on the ground and in space over a 
range of gravitational levels and with 
a variety of biological specimens; 
and (2) specialized equipment and 
advanced technologies are devel- 
oped to support life-sciences 
research on the ground and in space. 
The knowledge thus acquired is then 
integrated and disseminated by using 
information technology. 

In this report, new life-sciences 
research findings are presented; for 
example, new biochemical assays of 
bone collagen degradation, which 
are useful tools for monitoring bone 
mineral loss (a major biomedical 
problem of long-duration human 
spaceflight). Other research shows 
that both cerebral blood flow 
velocity and cerebral oxygenation 
decrease prior to fainting. This 
finding improves our understanding 
of the typical orthostatic intolerance 
and fainting observed in many 
astronauts following spaceflight. 
Additional work has focused on 
understanding the process of dual 
adaptation (or separate adaptations 
to two mutually conflicting sensory 
inputs) as it applies to the altered 
gravity environment. These studies 
should provide greater insight into 
the processing of altered gravita- 
tional inputs directly relevant to an 
astronaut's ability to function well 
in space. 

Technological advances 
described in this report have numer- 
ous applications in the medical and 
biomedical fields. The Virtual 
Environment Surgery Workbench, in 
combination with software originally 
developed for an Ames spaceflight 
experiment with rodents, has 



111 



attracted much interest from recon- 
structive surgeons, and this tool Is 
expected to decrease the training 
time of a skilled craniofacial 
reconstructive surgeon from 20 to 
10 years. In addition, this technology 
may soon offer doctors in the 
emergency room real-time, high- 
fidelity, three-dimensional informa- 
tion so that accurate diagnoses and 
potential reconstructive alternatives 
can be quickly determined. 

Neural net software develop- 
ment has resulted in another medical 
technology aimed at increasing the 
capability of neurosurgeons. The 
software enables a robotic probe to 
'learn" a patient's brain characteris- 
tics, making it easier for a surgeon to 
locate the exact position of a tumor 
after surgery is under way and the 
normal swelling of brain tissue 
complicates the process. 

In the ongoing effort to develop 
methods to monitor vital blood 
constituents, a prototype system that 
has application to astronaut health, 
space life sciences, and Earth-based 
medicine has been created. Concen- 
trations of ions in flowing blood 
plasma can now be measured 
by a computer-controlled data- 
acquisition system known as the 
Blood Flow Ion Analyzer. Liver- 
function support systems are direct 
beneficiaries of this technology, as 
there is a specific need to continu- 
ously monitor and control ion 
concentrations of processed blood 
plasma. 

A spinal compression harness 
designed as a potential aid to 
astronauts in combating back pain 
during spaceflight is also of interest 
to orthopedic therapists. The harness 
allows doctors to perform magnetic 
resonance imaging while positioning 



the patient in the characteristic 
supine position but with loads added 
to the spinal column simulating 
those that occur naturally in the 
upright posture. Such imaging 
capability will provide a greater 
range of diagnostic data for doctors 
who treat the millions of people who 
suffer back pain. 

Another recent Ames project 
involving telemedicine, Spacebridge 
to Russia, linked three U.S. hospitals 
and a university with two medical 
centers in Moscow via the Internet. 
Ames is responsible for developing 
the technical architecture as well as 
providing direction In setting up the 
Internet connections and the motion 
video consultation and video 
teleconferencing features. These 
capabilities have already enabled 
telemedicine consultations among 
all test users, and plans are under 
way to utilize the system as an 
educational tool for hospitals. 

As part of the HEDS goal to 
explore the solar system, Ames has 
actively pursued life-support meth- 
odologies relating to atmosphere 
regeneration, water purification, 
recycling from waste materials, and 
food production. The work has given 
birth to new concepts of reducing 
spacecraft power and weight 
requirements. 

Research in the unique environ- 
ment of space requires specialized 
equipment to operate within 
microgravity and assure that the 
research conducted does not pose a 
health or safety risk to the human 
crews. One such piece of equipment 
developed during the past year is the 
standard interface glovebox, which 
is now flying aboard the Mir station. 
This device was designed to provide 
an enclosed workspace that contains 



particulates, odors, and the kinds of 
chemicals and biological specimens 
required by life scientists. 

To help astronauts achieve 
maximum productivity during the 
particularly crucial first days of 
flight, researchers at Ames have 
been working on ways to reduce 
the space adaptation syndrome. 
Autogenic-Feedback Training (AFT) 
seems to be effective in controlling 
motion sickness under a variety of 
conditions, including spaceflight. 
The AFT exercise technique can also 
be used to control blood pressure 
and thus ameliorate postflight 
orthostatic intolerance. 

Lower-body negative pressure 
equipment specifically developed as 
a potential countermeasure for the 
debilitating effects of long-term 
human spaceflight has been modi- 
fied to provide a variety of loading 
levels on the lower body. In combi- 
nation with a treadmill, the equip- 
ment has applications both as a 
countermeasure to the atrophy of 
the musculoskeletal system during 
spaceflight and as a rehabilitative 
tool for treating patients suffering 
spinal trauma on the Earth. 

The use of a device that com- 
bines resistive exercise with aerobic 
training is described. Presently, 
cardiovascular exercise is empha- 
sized during Space Shuttle missions; 
however, research with the ''SX-1 
Variable Resistance Exercise Device'' 
suggests that astronauts may be able 
to maintain muscle size and strength 
while also maintaining bone density 
during spaceflight. 

It is vitally important that the 
results of HEDS research be quickly 
disseminated to the public, that is, 
the commercial sector and the 



112 



H f MAS E X P L R A r I \ A S D 
D E V E L F M E M Of S P A C E E ST ERP Rl S E 



educational and scientific communi- 
ties. To this end, life-sciences 
spaceflight data are being published 
in the scientific and popular litera- 
ture and archived in the National 
Space Science Data Center at the 
Goddard Space Center. Ames is also 
actively engaging teachers and 
students to promote science educa- 
tion of our nation's youth. 



113 



Biochemical Markers of 
Bone Metabolism in a Rat 
Spaceflight Model 

Meena Navidi, Jeanne Wren, 
Sara Arnaud 

Newer biochemical assays of 
bone collagen degradation products 
in urine are useful tools for monitor- 
ing bone loss (a major concern of 
spaceflight) and the effects of 
countermeasures to prevent bone 
loss in human adults. The results 
reflect the collective metabolic 
activities of bone resorption in the 
entire skeleton, a process that is 
activated by exposure to a head- 
down tilt, bed-rest spaceflight 
model. These techniques have yet to 
be applied to small-animal models 
of weightlessness. Through histo- 
logic methods, the decrease in new 
bone formation in the rat tibia during 
spaceflight is well documented. 
Recently, increases in osteoclast 
number and resorption were 
observed in highly localized areas 
in the proximal tibia. Ground-based 
models of juveniles of the same 
species show morphologic changes 
consistent with reduced bone 
formation in unloaded hind limbs 
with no evidence for increased 
resorption. The mature rat, exposed 
to the same ground-based model 
with unloaded hind limbs for 
28 days, reveals less bone mineral 
content in the femur, compared with 
ambulatory controls. This apparent 
bone loss is thought to be caused 
by increased resorption as well as 



depressed formation, but no histo- 
logic or biochemical measurements 
support this idea. 

To evaluate bone resorption, 
an assay that measures deoxypyri- 
dinoline (Dpd), a product of the 
breakdown of collagen crosslinks in 
bone was used. This assay is cur- 
rently regarded as one of the most 
specific and sensitive markers of 
bone resorption. The 24-hour 
excretion of Dpd in two-month-old 
juveniles was more than seven times 



840 



720 



^ 600 



a 

S> 480 

o 

S 

I 360 

c 

Q 240 



120 




JL 




Mature rats 



higher than in six-month-old mature 
animals (see the figure). This age 
difference reflects growth rates that 
average 2-3 grams per day in 
juveniles and are stable and essen 
tially zero in mature animals. 
Exposure to the spaceflight model 
for 14 days reduced Dpd excretion, 
whereas it remained unchanged in 
the control juveniles. In contrast, 
mature rats showed increases in Dpd 
excretion after 14 days exposure to 
unloading, as shown in the figure. 



Control D 
Space flight model ■ 



T 




Juvenile rats 



Fig. 1. The 24^hour excretion ofdeoxypyridinoline in the urine of juvenile 
and mature rats prior to and 14 days after exposure to a spaceflight model in 
which the hind limbs were unloaded. Note the higher excretions (p < 0.05) 
in the young animals than in the mature animals, and the opposite responses 
to the model at each age. 



114 



A s T K N A II T H K A L T II / Sclerice 



H i M A S E X P I R A T I S A S D 
D E V E L P M E S T OF S F A C E ENTERPRISE 



These results illustrate the 
marked, age-dependent differences 
in the response of the skeleton to a 
weightless environment. Results in 
the mature animals, who at six 
months of age have acquired 96% 
of the bone they will have in their 
lifetimes, are similar to the increases 
observed in human adults. There is 
an increase in bone resorption after 
unloading the skeleton, a situation 
favorable to bone loss unless 
accompanied by an increase in bone 
formation. Alkaline phosphatase, a 
circulating marker of bone forma- 
tion, was unchanged in this study. 
The opposite result, suppression of 
high resorption rates in juveniles, 
could be related to depressed growth 
hormone or a growth factor during 
exposure to the model. While it is 
clear that the adaptation of the 
skeleton to microgravity requires 
site-specific changes in the rates of 
formation and resorption in weight- 
bearing bones (to modify bone 
structure for new loads), there also 
appears to be a systemic response 
reflected in bone marker assays 
that quantifies overall skeletal 
metabolism. 

Point of Contact: M. Navidi 
(650) 604-5755 
mnavidi@mail.arc.nasa,gov 



Cerebrovascular 
Responses Prior to 
Fainting 

Kana Kuriyama, Toshiaki Ueno, 
Richard E. Ballard, 
Donald E. Watenpaugh, 
Suzanne M. Fortney, 
Alan R. Hargens 

Reduced orthostatic tolerance, 
commonly observed after space- 
flight, occasionally causes fainting 
during standing. Prefainting condi- 
tions are usually manifested by 
symptoms such as a sudden drop of 
heart rate or blood pressure that is 
followed by loss of consciousness if 
standing is continued. Although the 
cerebrovascular system may play an 
important role in postflight fainting, 
there have been fev^ reports con- 
cerning cerebral hemodynamics 
prior to fainting. This study was 
therefore undertaken to investigate 
cerebrovascular responses by 
measuring cerebral blood-flow 
velocity and cerebral tissue oxygen- 
ation prior to fainting, using lower- 
body negative pressure (LBNP) as an 
orthostatic stress. 

Seven healthy male volunteers 
were exposed to supine LBNP in 
steps of 10 millimeters Mercury 
every 3 minutes until prefainting 
symptoms were detected. Blood 
pressure and heart rate were mea- 
sured with an automatic finger-cuff 
device. Cerebral blood-flow velocity 



within the middle cerebral artery 
was measured with transcranial 
Doppler sonography. Cerebral tissue 
oxygenation was monitored using 
near-infrared spectroscopy over the 
forehead. 

During the initial stages of 
LBNP stress, when blood pressure 
decreases and heart rate is elevated, 
these cardiac responses are caused 
by blood pooling in the legs. But 
as LBNP continues, cardiac insuffi- 
ciency occurs prior to fainting. This 
study focused on the 2-minute 
period prior to the fainting endpoint 
when LBNP was stopped. In a 
typical subject, blood pressure 
continued to decrease gradually as 
fainting was approached (see the 
figure). Heart rate decreased rapidly 
in the final 30 seconds, signifying 
cardiovascular insufficiency just 
prior to fainting. 

The new finding of this study 
is that both cerebral blood-flow 
velocity and cerebral tissue oxygen- 
ation decrease prior to fainting. The 
decrease of cerebral blood-flow 
velocity can be explained as a 
decrease of cerebral blood flow or 
a dilatation of the middle cerebral 
artery. The decrease of cerebral 
tissue oxygenation can be explained 
as a decrease of cerebral blood flow 
or less likely, a decrease of cerebral 
tissue metabolism. Therefore, it is 
probable that cerebral blood flow 
decreases just prior to fainting. 



A S T R N A 



H K A 1. r 



/ Science 



115 



110 

I lool 

1^ 90 

■D E 80 

o ^-^ 

o 

ffl 70 

60 




W^ 






-120 -100 



-40 -20 




End 

point 


Time (sec) 





60 



Fig. 1. Blood pressure, heart rate, cerebral blood-flow (CBF) velocity, and 
cerebral oxygenation with emphasis on the final 2 minutes prior to lower- 
body negative pressure (LBNP)-induced fainting. Baseline values of each 
measurement (n) represent supine data obtained just before starting 
LB NP protocol. 



In conclusion, during LBNP- 
induced fainting, cardiac insuffi- 
ciency primarily occurs, resulting 
in the failure of cerebrovascular 
circulation. This study improves the 
understanding of postflight fainting 
in astronauts and may aid in the 
development of preventive methods. 

Point of Contact: A. Hargens 
(650) 604-5746 
ahargens@mail.arc.nasa.gov 



Chronic Exposure to 
Hyper-G Suppresses 
Otolith-Spinal Reflex in 
the Rat 

Nancy G. Daunton, 

Merylee Corcoran, Robert A. Fox, 

Li-Chun Wu 

As part of an ongoing effort to 
understand the neural mechanisms 
underlying adaptation to altered 
gravity, behaviors reflecting the 
function of gravity receptors (otolith 
organs) in the inner ear (for example, 
air-righting reflex, orientation) of the 
rat have been studied following 
chronic hyper-gravity (hyper-G) 
exposure and found to be consider- 
ably altered. In addition to the 
behavioral data, morphological 
studies conducted by Ross in 
conjunction with the behavioral 
studies have shown that there is a 
reduction in numbers of synapses on 
otolithic hair cells in the utricular 
maculae of animals exposed to the 
same chronic hyper-G conditions. 
These behavioral and morphological 
findings suggest that the sensitivity or 
gain in the otolithic system may be 



116 



A s T H N A M T PI K A 1. T li / Scieuce 



H i M A S E X P L R A T I S A S D 
D E y E L P M E S T OF S P A C E E S T E R P R i S E 



decreased following chronic 
hyper-G exposure. 

One way to evaluate the gain or 
sensitivity in the otolithic system is 
to use the ''free-fall" response, which 
is an otolith-spinal reflex elicited by 
a sudden brief vertical-linear accel- 
eration or drop. Using electromyo- 
graphic (EMG) techniques, this 
response has been demonstrated in a 
number of muscles in animals and 
humans. The earliest component of 
this response is dependent on intact 
otolith organs and does not habitu- 
ate over trials. 

The otolith-spinal reflex 
response to sudden free-fall was 
used to monitor sensitivity or gain 
in the otolithic system following 
chronic exposures to hyper-G. The 
investigators hypothesized that the 
amplitude of the early EMG response 
would be lower in animals exposed 
to hyper-G than in animals experi- 
encing only the normal 1 -G 
environment. 

A free-fall apparatus was 
developed that allows the animal, 
in a prone position, to be dropped 
suddenly over a 40-centimeter 
distance and then gently stopped by 
pairs of "soft springs." This device 
provides a sudden, vertical-linear- 
accelerative, free-fall stimulus to the 
animal being tested. Animals were 
instrumented with EMG electrodes 
in the gastrocnemius muscle of the 
hindlimb. The EMG response, along 
with acceleration of the animal, was 
recorded during 10 drops for each 
animal. The animals were tested 
within 3.5 to 9 hours following 
chronic (7 to 14 days) exposures to 
hyper-G produced by centrifugation 
on the Ames Research Center 
24 Foot-Diameter Centrifuge. 



Analysis of the data showed that 
in animals exposed to hyper-G, the 
majority of responses (61 %) to the 
free-fall stimuli were of very low 
amplitude (suggesting low sensitivity 
or gain), while normal, control 
animals had very few (13%) low- 
amplitude responses (see the figure). 
Furthermore, a significantly higher 
percentage of responses were low 
amplitude in animals exposed to 
2.8 G than in those exposed to 
2.0 G. This suggests greater loss of 
sensitivity or gain with centrifugation 
at higher G levels. In all cases 
animals exposed only to the rota- 
tional component of centrifugation 
had response amplitudes similar 
to those seen in normal control 
animals that were not exposed to 
centrifugation. 



The results of this study suggest 
that chronic exposure to increased 
gravitational conditions leads to a 
decrease in the sensitivity or gain of 
the gravity-sensing (otolith organ) 
portion of the vestibular system. 
Furthermore, a greater degree of 
suppression is produced by expo- 
sures to higher G levels. This 
decreased sensitivity may underlie 
the deficits in otolith-mediated 
behaviors seen following chronic 
hyper-G exposure and may be 
related to the reduced numbers of 
synapses found in otolith organs 
under the same conditions. The 
critical next question is whether the 
free-fall response will be increased 
following chronic exposure to 
microgravity. If that is the case, 



Free-Fali Response 

Gastrocnemius muscle 



Hyper-G 
(n = 4) 



Control 
(n = 4) 




Drop 



250 ms 



Average rectified EMG - R + 4 hrs 



Fig. 1. Average rectified electromyographic (EMG) response from four 
animals exposed to 2 G for 7 to 12 days and from four control animals, 
showing reduced amplitude of average response in hyper~G animals as 
compared with control animals. 



A s T R N A If T H K A 1. T \i I Science 



117 



important evidence will be obtained 
on how the otolith organ systenn 
is modified to subserve optimal 
vestibulo-spinal function (orienta- 
tion, postural control, locomotion) in 
altered gravitational environments. 

Point of Contact: N. Daunton 

(650)604-4818 

ndaunton@mail.arc.nasa.gov 



"Dual Adaptation" to 
Space-Related Sensory 
Rearrangements 

Robert B, Welch 

A series of studies is examining 
the process of ''dual adaptation" in a 
variety of atypical sensory environ- 
ments representative of those likely 
to be experienced by astronauts. The 
term ''dual adaptation" refers to the 
acquisition of separate, relatively 
independent adaptations to two (and 
perhaps more) mutually conflicting 
sensory environments resulting from 
repeated alternation between 
exposure and adaptation to one of 
the sensory environments and 
exposure and adaptation to the other 
environment. Evidence of dual 
adaptation is seen in a progressive 
decline in (1) the perceptual and/or 
behavioral interference that occurs 
at the initial transition between the 
two sensory conditions, (2) the 
amount of exposure necessary to 
produce asymptotic adaptation, 
and/or (3) the amount of adaptation 
acquired after a given amount of 
exposure. During FY96, three 
investigations of dual adaptation 
were completed. 



The first investigation, 
conducted by R. B. Post (University 
of California, Davis) and R. B. Welch 
(NASA Ames Research Center), 
measured perceived movement of 
a stationary visual stimulus during 
head movements, known as ''appar- 
ent concomitant motion" (ACM). 
ACM was measured both before 
and after exposure periods in which 
subjects performed voluntary head 
oscillations while viewing a spot that 
moved in conjunction with the head. 
During these intervals, subjects 
looked at the spot as It moved 
alternately in the same direction as 
the head was moving during either 
0.25 or 2.0 hertz head oscillations, 
and then in the opposite direction 
as the head at the other frequency. 
After a period of exposure, the visual 
stimulus came to be perceived as 
stationary only if it was caused to 
move in the same direction as that 
viewed during adaptation at the 
same frequency of head motion. 
Thus, these results confirmed the 
presence of dual adaptation of ACM 
in which the stimuli used to identify 
the prevailing sensory environment 
were those associated with head- 
movement frequency. 

in the second investigation, 
conducted by R. B. Welch 
(NASA Ames Research Center), 
B. Bridgeman and J. A. Williams 
(University of California, Santa 
Cruz), and R. Semmler (University 
of Clausthal, Germany), two experi- 
ments examined the possibility that 
the human vestibulo-ocular reflex 
(VOR) is subject to dual adaptation, 
as well as "adaptive generalization," 
which is the ability to adapt more 
readily to a novel sensory rearrange- 
ment as a result of prior dual- 
adaptation training. In Experiment 



One, subjects actively turned the 
head during alternating exposure to 

(a) the visual-vestibular rearrange- 
ment (target/head gain = 0.5), and 

(b) the normal situation (target/head 
gain = 0.0). These conditions 
produced both adaptation and 
dual adaptation of the VOR, but no 
evidence of adaptive generalization 
when tested with a target/head gain 
of 1 .0. 

Experiment Two, in which 
exposure to the 0.5 gain entailed 
externally controlled (that is, passive) 
whole body rotation, resulted in 
VOR adaptation, but no dual 
adaptation. As in Experiment One, 
no evidence of adaptive generaliza- 
tion was found. It was concluded 
that it was the absence of a salient 
discriminative cue in the passive 
rotation experiment that prevented 
dual adaptation from occurring. 

In the third investigation, 
conducted by R. B. Welch and 
N, Daunton (NASA Ames Research 
Center), R. Fox (San Jose State 
University), and M. Corcoran and 
L. Wu (NASA Ames Research 
Center), albino rats underwent four, 
7-day periods of continuous 2-G 
exposure, by means of centrifuga- 
tion, which alternated with four, 
7-day periods of normal gravity 
(1 G). Before and after each 7-day 
period, subjects were measured on 
(a) their righting reflex when 
dropped suddenly from an initially 
supine position into a vat of water 
40 centimeters below ("air righting"), 
and (b) swimming behavior after 
striking the water. Latency to the 
occurrence of air-righting was found 
to be subject to a form of dual 
adaptation. Specifically, animals 
that were repeatedly tested showed 
a significant improvement in mean 



118 



A s T R r^ A i: T Health / Science 



Human E x f l o r a t i o n a s d 
d e v e l p m e s t of s p a c e e s t e r p r i s e 



speed of air righting (+47%; 
p = 0.008), while control animals 
who received 2-G exposure, but 
were only tested on righting {and 
swimming) after the last of the 7-day 
exposure periods, showed only a 
marginally significant (p = 0.1 2) 
25% improvement in mean 
air-righting speed. 

Point of Contact: R. Welch 
(650) 604-5749 
rwelch@mail.arc.na$a.gov 




Virtual Environment 
Surgery Workbench 

Muriel D. Ross 

The research and development 
of a virtual environment (VE) surgery 
workbench was undertaken by the 
Biocomputation Center in FY96. This 
workbench will be used for training 
reconstructive surgeons and for 
patient-specific planning of complex 
craniofacial and other reconstructive 
surgeries. The virtual environment 
technology will be useful to the 
Human Exploration and Develop- 
ment of Space (HEDS) Enterprise for 
training astronauts for long-term 
missions in space. It will provide VE 
simulation capability during space- 
flight to help astronauts respond 
appropriately to unanticipated 
emergencies. 

The Immersive WorkBench 
(Fakespace, Mountain View, CA), 
is used for displays (shown in the 
figure (see Color Plate 20 in the 
Appendix)). This workbench is large 



and permits several individuals to 
see the visualization projected above 
the table top in three dimensions. 
Special glasses are required (see 
the figure). The images are greatly 
enlarged so that a novice can see 
details readily. The images can then 
be reduced in size to attain a more 
true-to-life size for further practice. 
This workbench will be used as 
interactive, VE tools are developed 
for training and planning purposes. 
Also, displayed images can be sent 
to remote sites over a network. The 
goal is to have interactive displays so 
that researchers and surgeons at 
remote sites can see identical three- 
dimensional (3-D) images and 
interact with them. 

In the Biocomputation Center, 
unique software has been developed 
that permits detailed, 3-D recon- 
struction of the face and skull of 
patients from computerized tomog- 
raphy (CT) scans. The Visible 
Human Dataset provided by the 
National Institutes of Health was 
used to reconstruct the face and 
skull of a 57-year-old female. In the 
reconstruction of the face, even tiny 
wrinkles around the mouth can be 
seen, and deeper wrinkles and the 
texture of the skin are realistic. 
Because of the level of detail 
possible, a laser scan to record the 
face prior to surgery is no longer 
needed. The images are rendered 
with high fidelity. 

The face and skull of an infant 
patient was generated from format- 
ted CT scans in about 20 minutes 
using the newly developed software 
and a Silicon Graphics Onyx RE2 
workstation. The infant had prema- 
ture closure of a skull suture on the 
right and of part of the anterior 
fontanelle (commonly known as the 



''soft spot''). The images of the skull 
clearly show the extent of fusion of 
the bones. Without surgical interven- 
tion, the right side of the head would 
not grow properly, the head and face 
would be deformed, and the brain 
would be compromised. Because of 
the high fidelity of the viewed 
images, surgeons have declared the 
software ready to use as a diagnostic 
tool. The goal is to have even more 
complex reconstructions completed 
within 15 minutes, or less, making it 
possible for doctors in emergency 
rooms to have a patient scanned and 
see the trauma in 3-D within min- 
utes. In addition, work is beginning 
that will make it possible to port the 
software to less expensive worksta- 
tions, promoting wider availability of 
the technology. 

A necessary part of the software 
development is the ability to cut out, 
to remove, and to insert pieces of 
bone elsewhere, then to computa- 
tionally restore the soft tissues so that 
the new face can be molded to the 
new skull for viewing prior to 
surgery. Software has been devel- 
oped to do this using a model grid 
system. The software will be incor- 
porated for use with an actual skull 
as the next step in software develop- 
ment. The software will be generally 
applicable, for example, in recon- 
structive surgery utilizing skin to 
recreate a breast in a mastectomy 
patient. 

The value of the training aspects 
of the technology should not be 
underestimated. It takes as many as 
20 years to train a highly skilled 
craniofacial reconstructive surgeon. 
If training in a simulator reduces this 
amount by half, 10 productive years 
will be added to a surgeon's profes- 
sional life. Furthermore, use of 



T \i C H N {) I, (; V A !• 1* I [ C ,\ T IONS T H [ M A N H K A 1. T W 



119 



simulator technology greatly reduces 
the cost of training. It is cheaper to 
practice on a computer than to 
participate in surgery on a patient 
and learn in this prolonged way. 
Also, new approaches to surgery 
can be quickly incorporated into a 
simulation to increase capability in 
new surgical techniques. 

The VE technologies that are 
being developed will reduce medi- 
cal costs, shorten the time necessary 
to train surgeons, and permit better 
preplanning of surgery. They also 
will improve the quality of life for 
the patient by producing a better 
outcome when surgery is first carried 
out. As these beneficial VE technolo- 
gies are transferred to the public, 
NASA also benefits from these 
applications that are useful to the 
HEDS Enterprise and to long-term 
spaceflight. 

Point of Contact: M. Ross 
(650) 604-4804 
mross@mail.arc.nasa.gov 



Noninvasive Estimation 
of Pulsatile Intracranial 
Pressure Using Ultrasound 

Toshtakt Ueno, Richard E. Ballard, 
lohn H. Cantrell, William T. Yost, 
Alan R. Hargens 

Exposure to microgravity causes 
a blood shift toward the head that 
may elevate intracranial pressure 
(ICP). As widely observed in patients 
with brain tumors or hemorrhage, 
elevation of ICP may cause head- 
ache, nausea, or vomiting. These 
conditions strongly suggest that 
elevated ICP in microgravity may 



cause or exacerbate similar symp- 
toms during spaceflight. In addition, 
recent clinical reports suggest that 
elevated ICP impairs regulation of 
blood flow in the brain and normal 



neural activity. Consequently, ICP is 
a critical parameter for understand- 
ing physiologic adaptation to 
microgravity. However, the magni- 
tude and time course of altered ICP 



-.03 |- 








-1 50 

DDI 1 












rPLL 














ICP - 




-.04 












40 


-5- -.05 


- 












O) 










~ 




(0 














% -.06 

> 


A At A 


A . k 1 


H h A i 


(\ Ai /I 


A A /I A r 


30 o> 


= -.07 


i 


h M 


\AU 


\nf\ 


hkk\h 


E 

E 


Q. 


n« K LI 


J \ lit 1 


\ . L M 1 1 r 1 


I ( \ M 


1 \ f \N \A \ J \ 




° -.08 


Y Ny 


wYvv 


vvVwSl 


'i 1* ,1 


JvYtI"-! 


20 & 


-J 






'i '* 


' * 1 * 1 






Q. 

0- -.09 


-1- n 1 1 
' 1 1 11 1 
' 1 1 111 


\ '» r ^ 


1 * ' 1 * * 1 , » < 
1 ' 1 ' * 1 . 1 1 


* ' 1 

' 1 1 


■ ; . .' ■ ;; ;'. ;; 

I * ' . • » 1 » . 


10 


-.11 


_ 1 \ t II 

\t M II 




1 ' 1 ' * 1 1 I 

.; li », ;i 1 




; '■' ' '. 'J 






> u u 


'- >' ", ' 


i' li V / \ 

1*1 1 




' • * 1 »' 




-.12 


1 t 


1 1 1 


1 1 i 1 1 


1 1 

















2 


4 6 


8 10 12 14 


16 18 20 






Time (sec) 







.006 |- 




10 






-•- PPLL 










-*- ICP 




.005 


— 






8 


N 


u 








X 






_ 




8, .004 


- 






^ 


jS 






_ 


6 5 


o 


- 








> 








I 


r .003 


- 




- 


E 

£ 


■S- 


- 




_ 


4 a 


Q 




► 




o 


-1 .002 


— 










t 


- 




1 y 


1 






1 ^ 


1 








t 


_ 


2 


.001 


y.ibfiWuiju^^^ 






01234567891 






Frequency (Hz) 







Fig. L Results of cadavers study. Typical waveforms in the pulse phase 
locked loop (PPLL) output and ICP are shown as a solid line and a dash line, 
respectively. ICP pulsations were generated manually by infusing saline into 
the ventricle at a rate of 1 cycle/second (1 hertz). The units of the PPLL 
output are voltages. The results of frequency analysis are superimposed. 



120 



T E C H N L C Y APPLICATIONS F H H M AN H K A L T 



h u m a s e x p l r a t i s a s d 
Development of Space Enterprise 



.004 


r ■ 








.003 


^ 


■ 


n : 


= 6 


_ .002 


r^ 








« .001 

i 


~ ■ 


1^ 


■ 




3 






^^""-^.^.^B 




S- -.001 


_ 




^^^--^ ■ 




j -.002 


- 




i"--^.- 




°- -.003 


- 




^^ 




-.004 


_ 




y = -6E-05x + 0.0012 




-.005 


- 




R^ = 0.749 

■ 




-.006 
-2( 


1 


1 


1 1 1 1 1 1 


1 


3 -10 





10 20 30 40 50 60 


70 








Tilt angle (deg) 





Fig, 2. Results of tilt study (n = 6). Amplitudes of pulsatile changes in the 
PPLL output are shown against the tilt angles, A linear regression line is 
superimposed. 



are still unknown, prinnarily because 
of the invasiveness of currently 
available techniques. 

A new ultrasound device to 
measure skull movements that occur 
with altered ICP has been developed 
and refined. The principle of this 
device is based upon pulse phase 
locked loop (PPLL), which enables 
the detection of changes in distance 
on the order of micrometers between 
an ultrasound transducer on one side 
of the skull and the opposite inner 
surface of the cranium. Although the 
skull is often assumed to be a rigid 
container with constant volume, 
many researchers have demonstrated 
that cranial bones move in associa- 
tion with changes in ICP. In the 
typical operation, the instrument 
transmits a 500 kilohertz ultrasonic 
tone burst through the cranium via a 



transducer placed on the head. The 
ultrasonic wave passes through the 
cranial cavity, reflects off the inner 
surface of the opposite side of the 
skull, and is received by the same 
transducer. 

In the first experiment, changes 
in cranial distance in three cadavera 
(ages 85-90) were measured with 
the ultrasound device, while pulsa- 
tile changes in ICP were generated 
by infusing saline into the lateral 
ventricle. To correlate the non- 
invasive measurement with ICP 
directly, a fiber-optic, transducer- 
tipped catheter was inserted into the 
intracranial space. In the second 
experiment, changes in cranial 
distance in six healthy volunteers 
(ages 18-31) were measured while 
they were placed in 60 degree, 
30 degree head-up tilt, supine, and 



10 degree head-down tilt positions. 
ICP was not measured directly. In 
both of the experiments, an ultra- 
sound transducer was taped securely 
to the lateral head above the 
right ear. 

In the cadaver study, frequency 
analysis (fast Fourier transformation) 
revealed that cranial pulsations were 
clearly associated with ICP pulsa- 
tions (see the first figure). In the tilt 
study, the magnitudes of cranial 
pulsations were linearly correlated to 
tilt angles as shown in the second 
figure. 

The ultrasound device has 
sufficient sensitivity to detect cranial 
pulsation associated with cardiac 
cycles. As previously reported, the 
amplitude of ICP pulsations is 
closely related to intracranial 
compliance. In other words, when 
intracranial fluid volume and 
pressure increase, arterial pulsation 
produces a higher amplitude ICP 
pulsation. Increased amplitude of 
ICP pulsations will be manifested by 
larger fluctuations in distance across 
the skull. Accordingly, by analyzing 
the magnitude of cranial pulsations, 
absolute estimates of ICP during 
spaceflight are possible. 

Point of Contact: A. Hargens 
(650) 604-5746 
ahargens@maiLarc.nasa.gov 



T K C H N [. {) C Y A I' 1' L [ C A T i N S T H T M A N H H A 1. T 



121 



MRI-Compatible Spinal 
Compression Harness 

Richard E. Ballard, 
Donald E. Watenpaugh, 
Iwane Mitsui, Klaus P. Fechner, 
Douglas S. Schwandt, Alan R. Hargens 

Sixty to eighty percent of the 
population of the United States 
experiences back pain during their 
lives. While noninvasive imaging 
systems such as magnetic resonance 
imaging (MR!) and computed 
tomography scanners have been 
used as diagnostic tools for studying 
back pain and orthopaedic injury, 
these imaging devices require 
patients to lie down, thus eliminating 
any effect of axial loading on the 
body that occurs in upright posture. 
As a result, these imaging systems 
often provide diagnostic data of 
limited use, particularly to disorders 
related to loading and unloading of 
the spine. 

The MRI-Compatible Spinal 
Compression Harness {a patent 
recently filed by NASA Ames) is 
designed to produce gravity-like 
axial compressive loads on the body, 
thus increasing the effectiveness of 
existing diagnostic imaging systems 
for detecting spinal disorders. In 
addition, the spinal compression 
device will help researchers study 
the physiology and biomechanics of 
intervertebral discs and causative 
mechanisms of back pain in patients 
on Earth and astronauts in space. 

The spinal compression appara- 
tus is lightweight, compact, and 
inexpensive, and fits into existing 
diagnostic equipment without costly 
modifications (see first figure). In its 
present configuration, the device 
consists of a nonmetallic footplate 



connected to shoulder pads, and 
alternatively to a head cap, by nylon 
webbing and elastic cords. Pulling 
the webbing through plastic buckles 
allows adjustment of cord tension, 
and thus compression level. The 
amount of compression is measured 
by nonmetallic force sensors that are 
currently being manufactured by a 
commercial partner. Alternative 
features of the device include a 
helmet-like headgear with adjustable 
elastic straps to provide neck 
compression, a saddle to distribute 
compression to the seat, and other 
attachments that allow static and 
dynamic compression of specific 
body segments or joints. A space- 
related benefit of the device includes 
the potential to compress the spine 



and alleviate back pain in astronauts 
during spaceflight, when spinal 
length increases between 4 to 
7 centimeters. 

In a recent pilot study, four 
healthy male subjects (age: 33-52; 
weight: 72-1 1 3 kilograms) were 
fitted with the spinal compression 
harness and placed in supine posture 
within an MRI scanner. Each sub- 
ject's spine was imaged without 
compression (control condition) and 
again at 25% and 50% body weight 
compression (the lumbar spine bears 
about 50% body weight in upright 
posture). Three of the subjects 
underwent an additional compres- 
sion level of 75% body weight (see 
the second figure). Longitudinal 
distance between T7/T& (thoracic 




Fig. 1. Percentage of body weight locally supported during upright posture 
(left) and supine compression in MRI (right). In upright posture, a gradient of 
weight-bearing exists from the head to the feet, whereas the harness for MRI 
compresses tissues uniformly between the sites of compression. Thus, loads 
are adjusted to the specific tissue to be scanned under load (for example, the 
lumbar spine typically experiences 50% body weight in upright posture). 



122 



T H C H N L G V APPLICATIONS TO H II M A N HEALTH 



H I If -1 \ E X P i R A T 1 \ A S D 

D t \ F. 10 p M E }, T Of Space E \ t f r p r i s f 



(a) 




(b) 

Fig. 2. Side view (a) and front view (b) of a spine during MRI with graded 
compression. Note the increase in curvature with increased load. This 
particular subject had a slightly bulging disc on the posterior side of L3-L4 
(third visible disc from the bottom) which was accentuated by compression. 
BW = body weight. 



level) and L5/S1 (lumbar/sacml level) 
discs decreased by 3.2 ± 0.8 milli- 
meters (mean ± SD) with 25% 
body weight compression, 
5.6 ± 1 .1 millimeters with 50% 
body weight compression, and 
6.8 ± 1 .2 millimeters with 75% body 
w^eight compression. The distance 
reduction was attri[)uted to 
dec reased disc thickness and 
increased spinal curvature. 

While further development and 
validation of this technology are 
necessary, these results indie ate that 
the spinal c omf^ression harness is 
capable of loading ihe spine effec- 
tively during supine MRI, and thus 
may be a valuable tool for studying 
effects of gravity-like loads on the 
spine. Previously this caf)ability has 
not l)een available to resc\irchers or 
clinicians. 

Point of Contact; A. Hargens 
(650) 604-5746 
ahargens@mail.arc.nasa.gov 



Near-Infrared 
Spectroscopy to Monitor 
Forearm Muscle 
Oxygenation 

Glta Murthy, Alan R. Hargens 

Muscle pain can be caused l)y 
repeated and constant exertion of a 
specific muscle. Such exertion may 
increase muscle pressure, reduce 
blood flow and muscle oxygenation, 
and cause muscle fatigue and pain. 
Low levels of prolonged muscle 
contraction of the forearm muscles 
are common in many work settings 
and may cause fatigue. Crew 
operations in the Life Sciences 



T i-: c H N I, {) (; v A v v i. [ c a r i o n s t 



K A I. i (I 



123 



Glovebox on the Space Station is 
one such work environnnent. The 
purpose of this study was to deter- 
mine whether forearm muscle 
oxygenation decreases significantly 
at low levels of contraction. Nine 
male and female subjects partici- 
pated in the study. Each subject was 
seated as shown in the first figure. 

Muscle oxygenation was 
measured noninvasively using a 
near-infrared spectroscopy tech- 
nique. The spectroscopy probe was 



placed over the forearm muscle and 
gently secured with an ace wrap. 
Subjects were asked to contract their 
forearm muscle maximally three 
times. After 1 minute of data collec- 
tion during relaxed posture, subjects 
contracted their forearm muscle 
once again for 1 minute, during four 
different load applications (randomly 
ordered). Loads were imposed by 
weights that were placed on the 
knuckles and allowed the subjects 
to work at 5, 1 0, 1 5, and 50% of 




Fig. 1 . Subject posture during the study. The inset shows the spectroscopy 
probe that measures muscle oxygenation. 



maximum voluntary contraction. At 
the end of the minute, the load was 
removed and a 3-minute recovery 
period followed. At the end of the 
protocol, with the spectroscopy 
probe still in place, subjects were 
required to exercise their forearms 
to exhaustion such that very little 
muscle oxygenation was recorded. 
This procedure is necessary to obtain 
a functional zero level for each 
subject. The data were normalized 
to a relative scale between the 
physiologic minimum (0%) estab- 
lished during the exhaustive exercise 
and the spectrophotometer output 
during relaxation (100%). 

After 35 and 40 seconds of 
contraction, muscle oxygenation 
leveled off at below baseline levels 
and remained at that level through- 
out the 1 -minute contraction period. 
Muscle oxygenation decreased from 
100"/;» during rest to 89 ± 4% (SE), 
81 ± 87n, 78 ± 8%, and 47 ± SJo at 
5, 10, 15, and 50"/) maximal forearm 
contraction, respectively (see the 
second figure). Muscle oxygenation 
decreased significantly at 10, 15, 
and 50'yo of maximal contraction 
compared to resting values. 

This study shows a significant 
reduction in muscle oxygenation 
even at levels as low as MYYo maxi- 
mal contraction. Muscle deoxygen- 
ation during muscle contraction may 
play an important role in the devel- 
opment of muscle fatigue and pain 
associated with repetitive tasks, 
especially in confined environments 
such as in a glovebox. There are no 
simple, easy to use instruments that 
can noninvasively measure muscle 
oxygenation in deep and superficial 



124 



T 1-: c w N I. (; v A v i* i. i c a t i t) n s t o 



H I M \ K H r f\ L T n 



H i M A V E \ P i R A r I S A A D 

D t y E I p M E \ T E Space E \ t e r p r i s e 




10 20 30 40 

Maximum voluntary contraction (%) 



50 



Fli^. 2. Muscle oxygen.Mion decreases with increasini^ lodd on the forearm 
muscle. 



working muscles. Thus, the spectros- 
copy device may provide an impor- 
tant tool to study muscle fatif^ue and 
pain and to improve the designs of 
workstations in space. 

Point of Contact: A. Hargens 
(650) 604-5746 
ahargens@mail.arc.nasa.gov 



BIONAl— Blood Flow Ion 
Analyzer 

John W. Mines, Christopher 1. Somps 

Liver support systems will 
benetil from continuous monitoring 
and control of the ion concentrations 
of processed blood plasma. To 
measure the concentrations of ions 
in flowing blood plasma, a proto- 
type, computer-controlled data 
acquisition system, the "Blood Flow 
Ion Analyzer" (BIONA 1), has been 
designed. BIONA 1 monitors plasma 
ion concentrations, displays them as 
a function of time, and saves the 
data to a file for later analysis. The 
sensors are miniaturized, catheter- 
type, ion-selective electrodes that 



have membranes selective to only 
one ionic species. Since the porcine 
liver cells decrease the plasma 
calcium over time, requiring the 
addition of Ca^+-enriched buffers, 
the initial ion of interest is Ca'^. 
Monitoring the pH and temperature 
of the blood plasma is also desired 
and can be implemented with future 
prototypes. 

To measure concentrations with 
ion-selective electrodes, a reference 
electrode that provides a cone entra- 
tion independent potential is 
needed. The potential of the ion- 
selective electrode (indicator 
electrode) depends on the concen- 
tration of the ion for which it is 
selective. The difference between 
reference and indie ator potential can 
then be related to the ion concentra- 
tion; a process called calibration. 
Ion-selective electrodes need to be 
calibrated before the first use and 
recalibrated after certain periods of 
time to minimize the amount of error 
generated by drift and changes in 
sensitivity (potential change per 
decade change of concentration). 

The BIONA 1 system prototype 
allows semiautomatic electrode 
calibration. Computer-controlled 
solenoid valves are used to tempo- 
rarily reroute blood plasma away 
from the sensor flow [3ath to a 
bypass flow path during calibration. 
One end of the sensor path, its inlet, 
is then connec ted to a syringe W'-ith a 
calcium chloride !)uffer solution, and 
the other end, to a waste container. 
This allows injection of buffer 
solutions for calibration w^ithout 
contaminating the blood plasma 
with buffer. After the buffers have 
been injected and calibration is 
finished, the inlet of the sensor path 



!•: C fl N' [- n (i ) A !• [' I, i C \ T [ \ S 



I 



I M A \ II K A L T l[ 



125 



is reconnected to the blood plasma 
flow while its outlet is still routed to 
the waste container. This results in 
the removal of buffer solution from 
the sensor path by the blood plasma 
flow. After a few seconds (flow-rate 
dependent), all buffer has been 
replaced by plasma. The sensor 
outlet is then switched back and the 
electrode tube is completely recon- 
nected to the blood plasma flow. 
The prototype is currently limited to 
a plasma flow rate of approximately 
100 cubic centimeters per minute. 

The electrode signal is amplified 
by a high-input impedance amplifier 
and acquired by a data acquisition 
PCMCIA card and a laptop com- 
puter. The same card also outputs 



digital signals to control (turn on/off) 
the solenoid valves. 

The solenoid valves require a 
1 2-volt DC power supply (at least 
6 watt) that is provided by a wall 
transformer. The amplifier circuit is 
powered by two AAA batteries for 
isolation purposes. To reduce the 
electromagnetic interference with 
the sensor signal, the box is copper 
clad to provide shielding. Shielding, 
however, is only effective if the box 
is grounded, which can be accom- 
plished by connecting the box with 
a ground wire to earth ground (for 
example, from an outlet). A sche- 
matic of the BIONA 1 system is 
provided in the figure. 



Blood plasma supply 



tnlet 



Bypass 



Sensors 



Outlet 




Laptop PC with 
PCMCIA DAQ-card 



Waste 



Fig. 1. Blood Flow Ion Analyzer (BIONA 1) system diagram. 



This technology development 
represented a diverse collaborative 
effort between researchers and 
technologists at the Center for 
Emerging Cardiovascular Technolo- 
gies (a National Science Foundation- 
sponsored program managed from 
North Carolina State University), 
and a team at the University of 
North Carolina at Chapel Hill and 
other affiliates in the Research 
Triangle, and at Case Western 
Reserve University in Cleveland, 
Ohio. A prototype Sensor Array for 
monitoring physiological ions in a 
hepatic bioreactor was demonstrated 
in a joint development activity with 
the Department of Surgical Researc h 
and the Liver Support Group at 
the Cedars-Sinai Hospital in 
Los Angeles, California. 

Point of Contact: J. Mines 
(650) 604-5538 
jhines@mail.arc.nasa.gov 



Intelligent Controller for 
Neurosurgery 

Robert Mah 

In preparation for neurosurgery 
to remove tumors, magnetic reso- 
nance imaging (MRI) scans are 
usually taken to locate the tumor 
and its positional coordinates. 
However, during the surgical 
operation to access the tumor, the 
brain swells and the location of the 
tumor shifts. As a result, the neuro- 
surgeon must rely on experience to 
search for the tumor interface using 
surgical tools and tactile feedback. 
(Brain tumors typically have a 



126 



T K C fi N 



,) L G Y A !' [= 1. I C A T IONS T H \' M A N H [■ A 1. T li 



H I >/ A \ E V P I. R A T I \ A S D 

D t: V F L F M E \ r OF Space E \ t e r p r i s e 




Fig, 1. Robotic neurosurgical hardware testbed. 



clitTorent density thdn normal brain 
tissue). During this procedure, an 
artery can accidentally be severed 
by the surgical tool which can lead 
to death. 

A simple robot that can 'learn" 
the physical characteristics of the 
brain may give surgeons more 
precise control of surgical instru- 
ments during delicate brain of)era- 
tions. The procedure utilizes a 
robotic smart probe that 'learns'' 
the brain's ( haracteristics by using 
neural net software. The probe, 
equipped with tiny sensors, enters 
the brain, gently locating the edges 
of tumors while preventing damage 
to critical arteries. Potentially, the 
neurocontrolled rol>ot, making slow, 
very precise movements during an 
operation, will be able to ''feel" 
brain structures better than any 
human surgeon. 

This research has contributed to 
the design of an intelligent robot that 
can improve the safety, accuracy, 
and efficiency of neurosurgery. This 
technology is expected to have 
direct spinoffs to the general field of 
surgery, in FY96, the NASA/Stanford 
Neurosurgical Computational 
Testl)ed was developed (see the 
figure). World-wide press coverage 
(major television stations, major 
newspapers, and radio stations) was 
received and an invention patent is 
pending. 

Point of Contact: R. Mah 
(650) 604-6044 
rmah@mail.arc.nasa.gov 



T K r II N I. (i V A !' r i. i c a t ions 



11 I W A N 



127 



Center for Health 
Applications of Aerospace 
Related Technologies 
(CHAART) 

Byron Wood, Louisa Beck, Sheri 
Dister, Brad Lobitz 

The Center for Health Applica- 
tions of Aerospace Related Tech- 
nologies (CHAART) was initiated at 
the beginning of FY95 by NASA 
Headquarters' Life and Biomedical 
Sciences and Applications Division 
(Code UL), within the Office of Life 
and Microgravity Sciences and 
Applications, The goal of CHAART is 
to promote the application of remote 
sensing (RS), geographic information 
systems (CIS), and related technolo- 
gies to issues of human health. Fiscal 
year 1996 began with the First Cyril 
Ponnamperuma Symposium on 
Remote Sensing and Vector-Borne 
Disease Monitoring and Control, 
held in Baltimore, Maryland, on 
November 28-30, 1995. The 
symposium, hosted by NASA and 
the Third World Foundation (TWF), 
was attended by national health 
representatives from Brazil, 
Cameroon, China, Indonesia, 
Malawi, Malaysia, Mexico, 
Myanmar, Namibia, Nigeria, 
Rwanda, and Zimbabwe. The goal 
of the symposium was to develop 
RS/GIS-based, health-related 
research proposals from Third World 
countries. The CHAART staff gave 
tutorials on RS and CIS and led 
panel discussions to define country- 
specific issues. 

As part of the new Memoran- 
dum of Understanding (MOU) 
between NASA and the Centers for 
Disease Control and Prevention 



(CDC), the CHAART staff worked 
intensively with CDC personnel in 
Atlanta, Georgia, to develop col- 
laborative projects. As a result of 
these discussions, the CHAART staff 
organized and taught two workshops 
on the uses of RS for public health 
applications. These workshops were 
conducted at CDC Headquarters in 
Atlanta (May 6-7, 1996), and at 
CDC's facility in Ft. Collins, Colo- 
rado (May 27-30, 1996). Recom- 
mendations for future NASA/CDC 
collaboration were generated at 
these events. As part of the NASA/ 
National Institutes of Health (NIH) 
MOU, the CHAART staff also spent 
the year working with researchers 
funded through the NIH. These 
researchers received supplemental 
funding through a joint NASA/NIH 
call, and used these funds to inte- 
grate RS/GIS into their ongoing 
research on disease. CHAART 
directly assisted Dr. James Maguire 
(Brigham and Women's Hospital at 
Harvard Medical School), who is 
studying leishmaniasis in Teresina, 
Brazil, and Dr. Rita Colwell (Univer- 
sity of Maryland), who is studying 
cholera in the Bay of Bengal. 

In other activities, CHAART also 
worked closely this year with TWF 
personnel to identify 10 visiting 
foreign scientists whose research or 
jobs involve human diseases. These 
scientists will spend two months at 
CHAART during FY97 learning how 
to integrate RS/GIS into their studies 
on human health. Two of CHAART's 
staff are on the Advisory Board of the 
Pan American Center for Earth and 
Environmental Studies. This center, 
established at the University of 
Texas, El Paso, as a Minority Univer- 
sity Research Center, was awarded a 
$6.5 million grant from the Mission 



to Planet Earth Enterprise to support 
environmental research in the 
Southwest. 

CHAART also provides RS/GIS 
support to Dr. Uriel Kitron (Univer- 
sity of Illinois, Champagne-Urbana), 
who is funded by NIH to study the 
distribution of Lyme disease in the 
Upper Midwest. 

Point of Contact: B. Wood 

(650)604-4187 

blwood@mail.arc.nasa.gov 



Telemedicine Spacebridge 
to Russia 

Steve N. Kyramarios 

NASA's newest telemedicine 
project, entitled Spacebridge to 
Russia, is a collaborative effort 
between NASA and the Space 
Biomedical Center for Research and 
Training in Moscow, and establishes 
a testbed to evaluate Internet 
telemedicine in both operational 
and educational settings. The project 
connects the Clinical Hospital of the 
Ministry of the Interior in Moscow 
and the Moscow State Medical 
School sites with Baylor Medical 
Center, Fairfax Hospital, Latter Day 
Saints Hospital, and Yale Medical 
School through the multicast back- 
bone (MBONE), 

The goal of this project is to 
provide medical telecollaboration 
and electronic patient record 
production and review through the 
Internet. The program will help 
NASA develop a permanent 
telemedical system that provides 
medical support for its astronauts 
and other personnel involved in the 



128 



T K C M N {) I. (I V A P \' 1. I C A T M.) N S T H V W A N H !■: A l T il 



H I M A S E X P L R A r I \ A s u 
D E ] F L P M F. S T OF S P A C E E \ T F R P R I S E 



joint Sf3ciceflight progmm with 
Russia. In addition, the experience 
gained from this effort will provide 
insight into the potential of medical 
practice over the National Informa- 
tion Infrastructure and the Glohal 
Information Infrastructure. 

This telemedicine testbed is 
between three clinical hospitals in 
the United States (U.S.) (Fairfax 
Hospital, Falls Church, Virginia; 
Latter Day Saints hlospital, Salt Lake 
City, Utah; and Yale University, 
New Haven, Connecticut) and two 
medical centers in Moscow (Clinical 
Hospital of the Ministry of Interior 
and the Space Biomedical Center for 
Training and Research at Moscow 
State University). UNIX-based 
Silicon Graphics Inc. multimedia 
w^orkstations are located at each site 
and are linked to the Internet. The 
workstations incorporate Internet 
tools including a graphical user 
interface (GUI) on the World Wide 
Web, and the MBONE toolset for 
video conferencing. It is estimated 
that several hundred clinical consul- 
tations will be conducted between 
the clinical sites in the United States 
and Russia. 

In addition to engineering the 
technical architecture, Ames 
Research Center has contributed to 
the project by providing technical 
leadership and implementation in 
Internet connectivity, motion video 
consultation, and video teleconfer- 
encing. These capabilities have 
enabled telemedicine consultations 
in a ^'grand round" format for unique 
case presentations. In addition, an 
educational program organized by 
Baylor Medical Center in Houston, 
Texas, is developing a series on 
recent advances in cardiology 
under the direction of Dr. Michael 



deBakey, the world famous heart 
specialist. The educational program 
includes live multicast video and 
web-based medical imagery. 
Clinical consultations are prepared 
at each site using a standard format 
that includes motion video, data 
transfer, still photography, and video 
teleconferencing capabilities. Cases 
are made available to the reviewing 
hospitals through a Web server. Each 
site [provides clinical coordination of 
the cases on a rotating basis. 

Point of Contact: S. Kyramarios 
(650) 604-4950 
skyramarios@maiLarc.nasa,gov 




Advanced Life Support/ 
Human Exploration and 
Development of Space 
Enterprise Activities 

Dick Lamparter, Mark Kliss 

The Ames Research Center 
(ARC) Advanced Life Support Branch 
carries out research and develop- 
ment of new technologies that will 
advance the human exploration 
and development of space. These 
activities will reduce life cycle costs, 
improve operational performance, 
promote self-sufficiency, and 
minimize expenditure of resources 
for future space exploration. Signifi- 
cant opportunities for Earth applica- 
tion of the developed technologies 
are available, including: 

Atmosphere Regeneration: Data 
on the influence of water on the 



adsorption of carbon dioxide and 
trace contaminants show significant 
reduction in capacity of sorbents and 
the possibility of elution back into 
the cabin environment. New materi- 
als and operating procedures will 
help mitigate this safety problem. 
New knowledge of adsorption 
processes has lead to a method of 
in situ resource utilization of oxygen 
for Mars missions. Experimentally 
validated, the process is likely to 
fly on the Mars In Situ Propel lant 
Production/In Situ Resource Utiliza- 
tion (ISPP/ISRU) Precursor experi- 
ment in 2001 . The data being 
developed on materials have given 
insight into process modifications 
that will reduce power and space 
requirements. This research program 
supports contamination aspects of 
flight experiments. 

Recovery of Resources from Waste 
Materials: Not currently recycled, 
human and food wastes are key 
elements for advanced life-support 
systems that approach self- 
sufficiency. Both Supercritical Water 
Oxidation (SCWO) and incineration 
processors have been developed ancJ 
are currently under evaluation. Key 
issues being addressed include waste 
preparation and process feed and 
product stream cleanup, with a 
special focus on acid gases and 
other trace contaminants produced. 
A prototype incinerator will soon be 
delivered to the Johnson Space 
Center (JSC) for evaluation in an 
extended manned chamber test. The 
U.S. Navy has expressed interest in 
SCWO technology for its potential to 
reduce hazardous liquids to carbon 
dioxide and water while ships are 
deployed at sea. 



P K (1 R [-: S S IN I M I' K V I \ (; S 



T \< \ \ li [, 



129 



Food production: Activity in this 
area is leveraged off prior research. 
In collaboration with the National 
Science Foundation, a unit to 
produce fresh food at the South Pole 
is being developed. The analysis of 
dietary requirements, crop mix, and 
multicropping in growth chambers 
has application to the Human 
Exploration and Development of 
Space (HEDS) objective of providing 
self-sufficiency. Experience in 
dealing with crews in remote 
environments is applicable to NASA. 
Information generated from this 
program is being formatted as part 
of the development of the JSC 
chambers. 

These activities form the basis 
for an important technology transfer 
to native communities in Alaska 
where health hazards exist because 
of poor rural sanitation. Code STR 
has established cooperative agree- 
ments with the state, university, and 
local communities to apply the 
NASA-developed technology and 
systems analysis capability toward 
improving health, through diets and 
treatment of water and solid waste 
streams. 

Point of Contact: M. Kliss 
(650) 604-6246 
mkliss@mail.arc.nasa.gov 



Mir Hardware— Stepping 
Stone to Station 

Bonnie P. Dalton, James Connolly, 
Gary Jahns, Paul Savage 

The Mir program has not only 
allowed the performance of biologi- 
cal experiments under long-term 
microgravity exposure (for example, 
90 to 145 days), but also has pro- 
vided a realistic view of experiences 
during the International Space 
Station program in terms of opera- 
tions and hardware preparations. 
In contrast to the Spacelab profiles, 
in which major flights occurred 
every 2 to 3 years, Mir flights and 
the associated planning, hardware 



and experiment development, and 
pre/postf light processing have 
occurred three times annually and 
have typically been initiated no 
earlier than 1 .5 years prior to 
launch. As a result, hardware has 
been developed very rapidly, with 
applications to interim flights a 
principal objective (until the 
International Space Station is fully 
functional). 

During the past 2 years, an 
insect habitat and a standard inter- 
face glovebox (SIGB) have been 
developed for use on Mir. The insect 
habitat was built with the assistance 
of Lockheed Martin Engineering and 
Sciences Company and other 
contract support. 




Fig, /. Microgrdvity glovebox for Mir 



130 



K {', K K S S IN 



I M i> \i V 1 N (; S p A c ]•: T h a v k i. 



H IM A S EXFLORATIOS A S D 
D E V E L F M E S T OF S F A C E E S T E R F R I S E 



The SIGB was designed to 
provide a fully enclosed workspace 
for the execution of in-flight science 
procedures. The SIGB utilizes sliding 
rails and latches adapted from the 
Johnson Space Center's standard 
interface rack (SIR), allowing it to fly 
in virtually any existing or future 
space platfornn. The work surface of 
the unit is approximately 1 square 
foot and the work volume is 
2.3 cubic feet. Bounded by a 
fiberglass shell consisting of five 
layers of epoxy resin, it is finished 
with white urethane for easy clean- 
ing. Airflow within the unit is 
modeled after a laminar flow hood. 
The air filtration system provides a 
net negative pressure of 0.1 to 
0.6 inch water with respect to 
ambient. All portions of the SIGB 
airflow system maintain negative 
pressure with respect to ambient in 
all operational modes. The filtration 
system has been designed to provide 
full containment of spills for specific 
chemicals (used in biological fixative 
preparations), including Campbell's 
Fixative, paraformaldehyde, and 
Russian variant solution. Access for 
operation is through attached 
gauntlet ports. The SIGB, ready for 
mounting in the Space Transporta- 
tion Service vehicle, is shown in 
the figure. 

The insect habitat, known as 
the Beetle Kit, was designed for a 
NASA-5 increment experiment 
entitled ''Effects of Gravity on Insect 
Circadian Rhythmicity" that will 
study the Tenebrionid beetle. The 
hardware incorporates an automated 
circadian timing system (CTS), Since 
previous spaceflights have shown a 
notable effect on the CTS of both 



vertebrates and invertebrates, it is 
anticipated that the ability to per- 
form long-term measurements on 
Mir will greatly enhance the scien- 
tific community's knowledge of this 
subject. The Beetle Activity Module 
(BAM), the hardware housing the 
insect and the timing system, is an 
adaptation of an existing item of 
Russian hardware with expanded 
capabilities- Housing 32 beetles, the 
BAM is self-contained, and includes 
both a treadmill and a timing system. 
Food consists of a small quantity of 
dry oatmeal. Movement of a beetle 
on the treadmill indicates the 
insect's activity during portions of 
its circadian rhythms. Data collected 
from the flight experiment and from 
ground-based centrifuge studies at 
Ames Research Center will help 
determine if the CTS exhibits 
adaptation during long-term expo- 
sure to altered gravitational fields 
and whether that exposure changes 
the sensitivity of the CTS to flight. 

In summary, Ames' participation 
in the Mir program resulted in the 
development of important hardware 
used for long-duration investigations 
and the methodology for rapidly 
building and qualifying the hard- 
ware. Additionally, Ames' participa- 
tion has served as a vital educational 
stepping stone toward ensuring the 
success of activities onboard the 
International Space Station. 

Point of Contact: P. Savage 
(650) 604-5940 
psavage@mail.arc.nasa.gov 



Wireless Network 
Experiment for Space 
Shuttle/Mir 

Richard Alena 

The Wireless Network Experi- 
ment (WNE) was a risk-mitigation 
experiment designed to demonstrate 
the feasibility of using wireless 
networks aboard the International 
Space Station. The customer was the 
Shuttle/Mir Science Program at the 
Johnson Space Center (JSC), which is 
under the Space Station Program. 
The WNE consisted of a set of three 
computers that could communicate 
via the wireless radio network 
system, as shown in the figure. A 
network monitor program graphed 
network throughput in real time as 
the computers were moved through 
the habitable volume. This experi- 
ment was successfully conducted 
onboard both the Space Shuttle and 
Mir Space Station during the docked 
space transport system-76/Mir 21 
flight. 

The current International Space 
Station data system baseline does 
not provide adequate support for 
payloads and onboard computer 
systems. Adding additional cabled 
network resources is prohibitive in 
terms of cost and weight. A wireless 
network allows portable computing 
resources to be brought up to the 
Space Station and installed with 
minimal effort and, at low-cost, 
support a significantly augmented 
data processing capability for 
payloads and crew. 



Pro <; r k s s in I m p k o v i k c S i' a c [■: T r a v i- i 



131 



SIA module 



28 VDC and 
chassis 
ground 





Radio 

frequency 

wireiess 

network 



Peripheral 

interfaces 

(unused) 




PCMCIA RF network cards 

and antenna assembly • 3 places 



Keyboard cable 

WNS 

keyl>oard 

module 



Trackball cable 

WNS 

trackball 



Fig. 1. Schematic of the Wireless Network Experiment (WNE) system. 



The WNE was conducted in 
March 1996, Demonstration of the 
next-generation (Web-based) data 
system concept to NASA and JSC 
astronauts and engineering managers 
was made in May 1 996. The post- 
fh'ght testing and preliminary report 
were completed in August 1 996. 

Point of Contact: R. Alena 
(650) 604-0262 
ralena@mail.arc.na$a.gov 



Engine Diagnostic Filter 
System 

Tarang Patel 

The project objective was to 
develop posttest data analysis tools 
for potential deployment as real-time 
and in-flight instruments for the next 
generation rocket engines for space 
vehicles. Analysis was made on the 
exhaust plume data collected from 
the Space Shuttle main engines. The 
analysis capability is to give indica- 
tions of engine erosion and its 
potential source inside the motor. 



The real-time analytic capability 
is unique in its ability to identify the 
principal eroding metal species 
under way in an engine during 
engine operation. Such health 
monitoring activity can be used to 
shut down an engine test prior to a 
potential catastrophic failure. This 
step could lead to a possible in-flight 
capability, under research, which 
would be used for onboard diagnosis 
for health management, and could 
represent a safety check for the 
space vehicle crew. A much more 
immediate benefit of the posttest 
analysis is that it provides engineers, 



132 



P K (; K K S S IN 1 M i> R V 1 N (I S V A C K 



H A \ K I 



H tun A S E X F L R A T I S AND 
D E V £ L P H E S r OF S P A C E E S T E R P S ! S E 



involved in engine tear-dov^n 
activity following a test, a real 
indication of whether such an 
expensive task is necessary. 

This project was the world's 
first use of plume spectroscopy to 
produce realistic, quantified esti- 
mates of metal erosion from a rocket 
engine. During recent routine testing 
of a main Space Shuttle engine at the 
Stennis Space Center, the analysis 
tools coupled to data from a diag- 
nostic instrument were able to 
determine metal erosion taking place 
and pinpoint the event of a cata- 
strophic failure. The methods used 
in the analysis were even able to 
extract silver and copper lines, 
despite the fact that they were 
swamped in the rich hydroxyl 
grouping (OH) structure of the 
spectrum. 

Point of Contact: T. Patel 
(650) 604-4721 
maumau@nas.nasa.gov 



Formal Lightweight 
Approaches to Vah'dation 
of Requirements 
Specifications 

Steve Easterbrook 

The objective of this research 
was to evaluate the use of formal 
methods as a tool for independent 
verification and validation (IV&V) of 
software requirements specifications 
(SRS) for spacecraft flight software. 
Formal methods allow an IV&V team 
to test SRS for properties such as 
consistency, completeness, safety, 
and correctness. The research 



focused on lightweight methods, 
which can be applied to small 
chunks of a specification without a 
substantial investment in translating 
entire specifications into mathemati- 
cal notations. 

A series of case studies was 
conducted, applying formal model- 
ing techniques to 'live' projects, so 
that the results can be fed back into 
the project in time to be of added 
value. The case studies were chosen 
in response to real needs on existing 
projects; for example, where an 
additional level of assurance of the 
correctness of the requirements was 
needed, over and above that obtain- 
able through existing methods. The 
case studies completed in the past 
year related to the International 
Space Station. A range of different 
formal methods was applied, 
including software cost reduction, 
the prototype verification system, 
and the model checker software 
process improvement network 
(SPIN). In each case the amount of 
effort required to apply the method 
was examined, along with the types 
of benefits gained. This evaluation 
was mainly qualitative, as the 
baseline metrics do not exist for 
detailed quantitative comparisons. 

The analysis conducted during 
the case studies provided a signifi- 
cant addition to the IV&V performed 
on the Space Station SRS. A major 
case study examined the require- 
ments for the fault detection, isola- 
tion, and reconfiguration (FDIR) for 
the main communications bus on 
the space station. This study 
involved translation of the original 
prose requirements into a formal, 
tabular notation, and analysis of the 
resulting model for consistency, for 



key static properties, and for correct 
dynamic behavior. The figure shows 
an analysis of the timing behavior of 
the FDIR requirements using the 
SPIN tool. 

The formal methods allowed 
the IV&V team to quickly identify 
numerous weaknesses in these 
requirements, including ambiguous 
use of terminology, missing require- 
ments, and Inconsistencies in the 
timing requirements. These problems 
would normally not be detected 
until later in the development 
process, at which point the costs to 
correct them are significantly higher. 

Results indicated that the main 
effort involved in applying the 
formal methods is in translating the 
prose requirements into the formal 
notation. However, this effort is 
consistent with the level of effort 
normally applied to performing 
IV&V on critical functions. In 
contrast, the ability to spot errors 
and omissions is greatly improved, 
as over half the issues identified in 
the formal analysis could not be 
detected during the requirements 
phase using existing methods. 
Hence, formal methods provide a 
cost-effective approach to analyzing 
requirements, provided they can be 
focused. 

Formal methods have not yet 
gained widespread acceptance by 
software practitioners. Part of the 
problem has been an emphasis on 
adopting a baseline formal specifica- 
tion, from which to prove that design 
and implementation are correct. 
This project has demonstrated that 
a more realistic approach is to use 
formal methods for small pieces of 
modeling, especially during the early 



P R (i R H S S IN [ M I' K \ I N (i S I' A C K T R A \ K L 



133 






Fig. 1 . A screen 
snapshot from 
the SPIN tooh 
showing the 
FDIR system 
(center) 

interacting with 
other subsystems 
(the clock, a test 
harness, and the 
hardware it 
controls). This 
analysis allows 
the analyst to 
confirm that the 
software will 
behave in the 
intended fashion 
even before it is 
written; that is, 
based solely on 
a statement 
of the 
requirements. 




r^-^^^^^^jJi^^^S^^ • "*^'**''*'*"'^**''^ ^ ' -^^"^ ^^" mt^. 3f liftlft^T^llflft lIM^^^^ I . ; ■ ■ l il 



m 












ii 



II 



< %m\ 



134 



P K r, im: S S IN 



M [' K V I N (; 



S V A C !■: T K A \' F L 



tt I M AS E X P I R A T I S A > D 
D E V E 10 P M E S T OF S P AC E E S T E R P R I $ E 



requirements stages, to answer 
questions that cannot be addressed 
\n other ways. 

Point of Contact: S. Easterbrook 
(304) 367-8352 
S.Easterbrook@ivv.nasa.gov 




Autogenic-Feedback 
Training as a Potential 
Treatment for Postflight 
Orthostatic Intolerance 

Patricia S. Cowings, 
William B. Toscano 

The goal of this research is to 
determine if human subjects can 
learn to voluntarily control blood 
pressure increases during a gravito- 
inertial stimulus (for example, tilt 
table), and to determine the mecha- 
nisms by which such control is 
effected. The training method used, 
Autogenic-Feedback Training 
Exercise (AFTE), was initially devel- 
oped as a treatment for motion 
sickness. It involves teaching 
subjects to control voluntarily up 
to 20 of their own physiological 
responses simultaneously. In support 
of the present study, cardiovascular 
dynamics (derived from impedance 
cardiography) were measured and 
displayed to test participants in real 
time. These measures included: 
cardiac output, stroke volume, and 
an estimate of vagal tone. Tests are 
in progress; this report outlines the 
results obtained to date. 

Baseline data were collected on 
each subject during supine, head-up 
tilt, and head-down tilt conditions. 



Subjects were given two tests, each 
50 minutes long, one test before and 
one test after AFTE. Physiological 
data were continuously recorded 
during both tests with the Autogenic- 
Feedback System-2 and impedance 
cardiography. The protocol for these 
tests is outlined as follows: 

1 . Supine resting 1 minutes 
baseline 

2. Head-up tilt at 10 minutes 
50 degrees 

3. Supine resting 10 minutes 
baseline 

4. Head-down tilt at 10 minutes 
-30 degrees 

5. Supine resting 10 minutes 
baseline 

Each subject received 6 hours 
of AFTE, which was administered in 
twelve 30-minute sessions (with 
subjects sitting; that is, nontilt 
conditions). Following a 6-minute 
pretest baseline, each 30-minute 
training session was divided into 
ten 3-minute trials, during which 



subjects were instructed to alter- 
nately increase and decrease 
response levels (for example, heart 
rate accelerations and decelerations, 
peripheral vasodilatation and 
constriction, etc.). A 6-minute resting 
baseline was also collected after 
each session (total 42 minutes). 

The purpose of these training 
sessions was to provide subjects 
with the ability to recognize bodily 
sensations associated with both 
increases and decreases in their 
physiological response levels and, 
with practice, to improve their skill 
in controlling these responses. 
Subjects eventually learn to maintain 
their physiological response levels at 
or near their own resting baseline 
levels and improve their tolerance 
to environmental stressors (for 
example, tilt table, conditions of 
microgravity, etc.). 

The first figure shows a subject 
in the head-down tilt position. The 




Fig, 1. Subject during 30-degree head-down tilt. 



A s r R N A V T H K A I, T n / C u tui i er m e a s N v e s 



135 



Cardiac Output 
Subject #1 



Before AFT 
After AFT 




I i I I I I ' ' I t ' I ' ' I 1 t I ' ' ' 
10 19 28 37 46 

Minutes 



100 



90 



40 



Heart Rate and Blood Pressure* 
Subject #1 



♦ ♦ ♦ ♦♦♦ 



-| 170 



140 




» I i \ \ L 

10 19 28 

Minutes 



I L- 

37 46 



50 



Btood pressure was taken atter AFTE oniy. On the bottom ciiart, 
diamonds represent systolic blood pressure and squares 
Indicate diastolic blood pressure (read on right axis). 



Fig. 2. Physiological responses to tilt before and after AFTE. 



second figure shows physiological 
responses of cardiac output (top 
chart) and heart rate and blood 
pressure (bottom chart), before and 
after AFTE. This subject made a 
significant increase in both heart rate 
and cardiac output in response to 
head-up tilt (minutes 1 1 through 20) 
after eight AFTE sessions. There was 
no observable change in cardiac 
output during tilt before training. 

The third figure shows the 
physiological data of one subject 
during AFTE sessions four (after 
almost 2 hours of training) and eight 
(after almost 4 hours of training). The 
data shown include finger pulse 
volume, respiration rate, heart rate, 
skin conductance level, cardiac 
output, stroke volume, central 
volume (actually thoracic imped- 
ance), and an estimate of vagal tone 
(calculated by Fast Fourier Transform 
in real time). Following the 6-minute 
baseline, subjects were instructed to 
produce arousal responses (constric- 
tion of finger blood volume, heart- 
rate and skin-conductance-level 
increases) but not to modify respira- 
tion or to contract muscles. More 
consistent control of finger pulse 
volume was apparent during session 
eight than during session four— 
as the subject makes the desired 
response at 3-minute intervals. 
Respiration was recorded to show 
that the subject was neither hyper- 
nor hypoventilating (by session 
eight) to influence other response 
changes. By session four, both heart 
rate and skin conductance level 
show evidence of learning that 
remained consistent through session 
eight. Note, however, that learned 
control of cardiac output, stroke 
volume, and vagal tone was not 



136 



Astronaut H f. a l t h / Countermeasures 



H I M A S E X P L R A T I S A S D 

D E\ iio p M EST OF Space E m e r p r i s e 



Subject #1 



-AFT #4 
■AFT #8 



Respj ration rate 




17 21 25 
Minutes 



7 10 13 16 19 22 25 28 31 34 37 40 
Minutes 



Fig. 3. Physiological responses during AFTE sessions four and eight. Note: Following a 6-minute resting baseline, 
subjects are instructed to produce alternating arousal (that is, constriction of blood flow to the fingers, heart-rate and 
skin-conductance-level increases) and relaxation responses (that is, vasodilation of fingers, decreases in heart rate 
and skin conductance level) at three 3-minute training trials, showing for the first time that these parameters may be 
subject to learned, voluntary control. 



A s i H N A 1 T H H A L T li / C tnil CT m e (i s u r es 



137 



apparent until session eight, suggest- 
ing that AFTE practice is associated 
with observed response changes. 
Centra! volume was also modulated 
with instruction (that is, at 3-minute 
intervals) more consistently after 
eight sessions. 

These data represent the first 
indication that these measures of 
cardiac dynamics are subject to 
voluntary control through learning. 
They show that AFTE methods may 
yield a powerful, potential counter- 
measure for postflight orthostatic 
intolerance that may supplement 
existing exercise routines utilized by 
crewmembers. 

Point of Contact: P. Cowings 
(650) 604-5724 
pcowings@maiLarc.nasa.gov 



date has lacked sufficient loads to 
maintain preflight musculoskeletal 
mass. Although treadmill exercise 
with bungee cords {about 2 hours 
per day) is the most common 
exercise for cosmonauts during long- 
duration Mir missions, biomechani- 
cal loads on musculoskeletal tissues 
of the lower body are only about 
60 to 70% of those present on Earth. 
Theoretically, an integrated counter- 
measure for extended exposure to 
microgravity should combine high 
loads on the musculoskeletal system 
with normal regional distributions of 
blood pressure and stimulation of 
normal neuromuscular locomotor 
patterns. 

Lower-body negative pressure 
(LBNP) exercise is an integrated 
countermeasure that may prevent 



bed rest- and microgravity-induced 
deconditioning by simulating 
gravity. (See the figure.) The LBNP 
exercise device consists of a verti- 
cally oriented treadmill within a 
partial-vacuum chamber that is 
sealed around the subject's lower 
body. For use on Earth, supine 
subjects are supported by a suspen- 
sion system of cables and pulleys. 
Suction pressure is used to pull 
blood and other body fluids away 
from the heart (thus reproducing 
normal upright blood pressure 
gradients) while simultaneously 
creating a gravity-like footward 
force. 

In a recent study, it was hypoth- 
esized that daily LBNP exercise 
maintains cardiovascular and 
musculoskeletal conditioning during 



Exercise for Long- 
Duration Spaceflight 

Donald E. Watenpaugh, 

Richard E. Ballard, 

Karen |. Hutchinson, 

Jaqueline M. William, Andrew C. ErtI, 

Suzanne M. Fortney, Lakshi Putcha, 

Wanda L. Boda, Stuart M. C. Lee, 

Alan R. Hargens 

Adaptation to microgravity leads 
to cardiovascular and musculoskel- 
etal deconditioning to normal Earth 
gravity. Specific problems include 
muscle atrophy, bone demineraliza- 
tion, reduced neuromuscular 
coordination, and postflight reduc- 
tion of orthostatic tolerance and 
upright exercise capacity. Presently, 
exercise protocols and equipment 
for astronauts in space are unre- 
solved, although recent calculations 
suggest that all exercise in space to 



(Earth gravity) 



1 


1 


^ 


ffl 



(LBNP suction force) 



Fig. 1. Treadmill reaction force to generate body weight in supine posture 
equals the product of the cross-sectional area of the flexible waist seal and 
the suction pressure within the LBNP chamber With exercise, inertial 
reaction forces against the treadmill are added to the static force. 



138 



Astronaut H li a l t h / Countermeasures 



H i M AS E X P i R A T I .V A \ D 

D E y E I p M E s T OF Space E s t e r p r i s e 



6-degree head-down bed rest. Seven 
healthy male volunteers acted as 
their ov^n controls, such that their 
responses to 2 weeks of bed rest 
with daily supine LBNP treadmill 
exercise (40 minutes of walking and 
running at 1 .0 to 1 .2 body weights of 
footward force) were compared to 
their responses to 2 weeks of bed 
rest with no daily exercise. The 
2-week bed rest sessions were 
separated by 10 weeks. 

Forty minutes per day of LBNP 
exercise preserved upright exercise 
capacity during 2 weeks of bed 
rest. Subjects' time to volitional 
exhaustion during their individual- 
ized treadmill tests decreased 
1 .72 minutes on average {]()%, 
p < 0.05) after bed rest with no daily 
exercise; daily LBNP exercise during 
bed rest maintained exercise toler- 
ance time at pre-bed rest levels. 
Daily LBNP exercise also maintained 
peak upright oxygen consumption 
(VO2) at pre-bed rest levels (pre-bed 
rest: 59.5 ± 3.2 milliliters/minute/ 
kilogram; post-bed rest: 56.4 ± 3.4). 
Mean peak VC)2 decreased sig- 
nificantly from 57.6 ± 2.6 to 
49.8 + 1 .5 milliliters/minute/kilo- 
gram after bed rest with no exercise. 
Sprint speed from a standing start 
was maintained at pre-bed rest levels 
when daily LBNP exercise accompa- 
nied bed rest (pre-bed rest: 
5.5 ± 0.2 meters/second; post-bed 
rest: 5.2 ± 0.3). However, bed rest 
without daily LBNP exerc ise signifi- 
cantly reduced sprint speed from 
5.5 ± 0.2 to 4.6 ±0.3 meters/ 
second, below pre-bed rest control 
levels. 

In conclusion, LBNP exercise 
improves upon current spaceborne 
exercise technologies by supporting 



cardiovascular and musculoskeletal 
fitness while reducing the duration 
of astronaut exercise sessions. 

Point of Contact: A. Hargens 
(650) 604-5746 
ahargens@mail.arc.nasa.gov 



Keiser SX-1 Variable 
Resistance Exercise Device 

Jennifer Pedley, Anthony Artino, 
Richard Ballard, Alan R. Hargens 

Muscle atrophy and decreased 
bone density are problems associ- 
ated with extended exposure to 
microgravity. Therefore, it is vital for 
astronauts to exercise during long- 
duration spaceflight to reduce these 
adverse physiological changes. In 
collaboration with Ames Research 
Center and Palo Alto Veterans 
Administration Rehabilitation 



Research and Development Center, 
the Keiser SX-1 resistance exercise 
device was developed to prevent 
muscle atrophy and bone weakness. 
Two pneumatic pistons provide 
variable and high resistance for the 
lower-body exercises, whereas a 
single piston allows resistance 
exercises for the upper body and 
spine. When performing the exer- 
cises, the user is secured by a 
harness to provide muscle isolation 
and proper positioning (see the first 
figure). The device includes an 
adjustable monitor that provides 
feedback information such as force, 
work, and power. Other features 
include strength and cardiovascular 
modes represented on the monitor. 
In the strength mode, the monitor 
displays the number of repetitions 
and the angle of the upper-body 
exercise levers. The cardiovascular 
mode conveys to the user repetitions 
per minute, total time of the exercise 




Fi^. 1. The Keiser SX-1 in an untolcied confip^unitlon. 



^ -^ ■'■ f^ f> ^^ ■> !■ i" H K A I. r fi / Coinilenucdsures 



139 



for upper and lower body, and 
graphic indicators to represent the 
range of motion. The Keiser SX-1 
is designed to fit into a standard 
Space Station rack. The total weight 
of the device is 1 20 kilograms 
(265 pounds), but the weight can be 
reduced to 80-100 kilograms using 
aluminum. 

All exercises are performed in 
one or two positions by using the 
four buttons located on the 
handgrips. Two thumb buttons on 
each handgrip are capable of 
adjusting (1 ) the amount of resis- 
tance to the upper and lower 
extremities; (2) the leverage from a 
pulling to pushing exercise (that is, 
pulldown to a chest press); and 
(3) the range of motion for arms 
from to 70 degrees (that is, row to 
shoulder press exercise). Maximum 
and minimum resistance values for 
the lower-body range from 1 ,500 to 
100 Newtons and for the upper body 
from 1 ,000 to 50 Newtons. At any 
designated time, the user can adjust 
the resistance and exercise, 

A user can perform resistance 
exercises from supine and prone 
positions. These exercises include: 
(1 ) bilateral chest press for strength- 
ening the pectoralis major, anterior 
deltoids, and triceps; (2) bilateral 
row for strengthening the latissimus 
dorsi, rhomboids, and biceps; 

(3) bilateral shoulder press, supine 
and/or prone position for strengthen- 
ing the deltoids and triceps; 

(4) bilateral lat pull-down, supine 
and/or prone position for strengthen- 
ing the latissimus dorsi and biceps; 

(5) bilateral or unilateral leg press for 
strengthening the gluteals, quadri- 
ceps, hamstrings, calf muscles, and 



lower back postural muscles; and 
(6) bilateral or unilateral hip flexion. 
In addition, many of the above 
exercises produce spinal loading that 
may help reduce back pain due to 
spinal lengthening in microgravity. 

In a comparison evaluation with 
a standard supine leg press (SLP) 
device, the SX-1 demonstrated 
similar thigh muscle electromyo- 
graphic (EMG) activities during 
concentric and eccentric exercises 
(see the second figure). With regard 



to the calf muscles, gastrocnemius 
EMG activity was greater with the 
SX-1 than the SLP, while soleus EMG 
activity was similar or slightly lower 
with the SX-1 than the SLP. These 
results indicate that the Keiser SX-1 
is capable of loading thigh and calf 
muscles in a manner similar to a 
conventional leg press device used 
for ground-based exercise. 

Presently, cardiovascular 
machines are emphasized on the 
Space Shuttle; however, strength 



100 



D SX-1 



o 

s 

UJ 




Con 



Ecc Con Ecc Con Ecc 



Vastus medialis Vastus lateralis Rectus femoris 

denotes significant difference between SX-1 and SLP (p < 0.05). 



Fig, 2. Average thigh muscle electromyographic (EMG) activity in 11 
subjects for the concentric (Con) and eccentric (Ecc) phases of contraction 
during a 1 0-repetition maximum leg press exercise using the Keiser 5X- 1 
and conventional supine leg press (SLP). In general, EMC activities were 
similar between the SX-1 and SLP, with the exception of concentric 
contractions of the vastus medialis and rectus femoris. This difference is 
probably related to a lesser load placed on these muscles using a variable 
pneumatic resistance for the 5X- / as compared to a constant weight 
for the SLP 



140 



\ j^ T K N A I T H K A L T H / Co tin tev ffi e ci s u v es 



H i M AS E X P L R A T I A A S D 
D F\ E L F M E S T OF S P A ( E E S T E R P R I S E 



training devices are an exception. 
Exercising on a variable-resistance 
device during spaceflight may be 
effective in decreasing the detrimen- 
tal physiological effects caused by 
exposure to microgravity. KeiserSX-1 
has the ability to combine cardiovas- 
cular and resistance training as well 
as spinal loading during exercise. 
With an appropriate high-resistance 
training protocol, users may be 
able to maintain muscle size and 
strength as well as bone density. The 
Keiser SX-1 represents an important 
advance toward this goal. 

Point of Contact: A. Hargens 
(650) 604-5746 
ahargens@mail.arc,nasa.gov 



Dehydration at Airline 
Cabin Altitude 

John E. Greenleaf, Peter A. Farrell, 
Helmut Hinghofer-Szalkay 

A contributory factor for the 
jet-lag syndrome in humans is total 
body dehydration, or hypohydration 
(see first figure), which is manifest 
by a reduction in plasma volume 
(hypovolemia). Exposure to pro- 
longed chair-rest confinement and 
possibly to moderate altitude (alt.), 
as occurs in passenger airline cabins, 
could facilitate this dehydration. 
The purpose of this study was to 



determine if rehydration drinks 
(AstroAde (AA), Performance 1 (PI), 
or water (H2O)) could ameliorate 
the hypovolemia in 10 young men 
(21 to 30 years old, weighing 62 to 
101 kilograms) sitting (with light 
activity) for 12 hours at 2800 meters 
(9184 feet, 539 millimeters Mercury) 
in an altitude chamber. 

After the subjects rested for 
10 hours in the chamber and 
consumed three normal meals 
(2850 kilocalories, 1400 milliliters 
water), their plasma volume 
decreased significantly (p < 0.05) 
by 9.0% (AA, alt.), 6.2% (PI , alt.), 
7.4% (H2O, alt.), and 9.0% (H )0, 



































Confinement 
decondilioning 






Cabin environment 


Anxiety 












1 


i 1 






1 










Caffeine 
intake 


Increased blood 
pressure 

Increased heart rate 


1 — 


Reduced exercise 
sitting body position 










1 ^ 




1 














T T 




Increased 
urinary output 


Decreased fluid 
& food intake 




















AIco 


K^i 1- 


hol p 






' r 






i 


^ 


intake | 










Lower 

extremity 

edema 






Low humidity 

reduced p02 

increased airflow 






















+ 






\ 








Increased body 

water loss 
(dehydration) 






Increased insensible 

& respiratory 

fluid loss 




^ 








♦ 












Hemoconcentration 
(Hypovolemia) 












— 

































Fig. /. Factors affecting in-flight hypovolemia. 



A s r H N A [' T H [■: a 1, t h / Coini/ 



er measures 



141 



o 

X 



^ -6 

E 

3 
O 

i -8 

(0 

a 
Q. 




-10 



-12 



_1_ 



(8.3)* 



(2.8) 



A AstroAde - Attitude 
• Performance 1 - Altitude 
A Water - Altitude 
O Water -Ground 



A (-0-9) 
^ (0.8) 



_L 



Sitting - 

I 



_L 



10 



12 



2 4 6 8 

Time (hrs) 
' Significant (p < 0.05) decrease at hour 10, and significant percent 
increase at liour 12 (in parentheses). 



ground). (See second figure.) The 
men then drank AA, PI , or H2O 
(12 milliliters/kilogram body weight, 
948 milliliters at 16 degrees Centi- 
grade). At 90 minutes (hour 1 20 after 
the 30-minute drinking period), 
plasma volume had increased by 
8.3% (AA, p < 0.05), 2.8% (PI, not 
significant), -0.9% (H^O, alt., not 
significant), and 0.8% (H2O, ground, 
not significant). 

These data indicate that signifi- 
cant dehydration and hypovolemia 
occur with adequate food and fluid 
intake during moderate (1 2-hour) 
confinement (sitting) at moderate 
altitude. Because plasma volume 
also decreased on the ground, it 
appears that confinement, and not 
the moderate hypoxia, causes the 
hypovolemia and hypohydration. 

Point of Contact: A. Hargens 
(650) 604-5746 
ahargens@mail.arc.nasa.gov 



Fig. 2. Mean percent change in plasma volume at altitude and on the ground. 



142 



A s T R N A i T H K A 1. T H / Co uu term eci s u r es 




9fi33 Z3_5G0 G37-3 -p5t>i 






#■ 




M I s s I \ i n P I A \ F T Earth E s t e r p r i s f 







Overview 

NASA's Mission to Pianet Earth 
(MTPE) studies the total Earth 
environment^atmosphere, ice, 
oceans, land, biota, and their 
interactions — to understand the 
effects of natural and human- 
induced, near-term changes on the 
global environment and to lay the 
foundation for long-term environ- 
ment and climate monitoring and 
prediction. Ames Research Center 
supports the MTPE Enterprise by 
conducting research and developing 
technology to expand the knowledge 
of Earth's atmosphere and ecosys- 
tems. This objective is also one of 
the goals of NASA's astrobiology 
research and technology efforts, 
which are led by Ames. A comple- 
mentary objective is to apply the 
knowledge gained to practical, 
everyday problems and to transfer 
the technology and knowledge to 
users outside NASA. Numerous 
research and technology efforts were 
accomplished during FY96, and the 
results address the following goals of 
the MTPE Enterprise: 

• Expand scientific knowledge of 
the Earth's environmental 
system. 

• Enable the productive use of 
MTPE science and technology. 

• Disseminate information about 
the Earth's environmental 
system. 

The research is particularly 
concerned with atmospheric and 
ecosystem science and with bio- 
sphere/atmosphere interaction. Key 
components of the research include 
the study of physical and chemical 
processes of biogeochemical 
cycling; the dynamics of terrestrial 



and aquatic ecosystems; the chemi- 
cal and transport processes that 
determine atmospheric composition, 
dynamics, and climate; and the 
physical processes that determine 
the behavior of the atmosphere on 
the Earth and other solar system 
bodies. Of special interest are the 
development and application of 
new technologies required to bring 
insight to these science topics. 
Significant contributions were made 
in three areas: analysis of critical 
gaseous emissions, development of 
models and instruments, and 
radiation research. 

Research highlighted in this 
report includes focus on imf)ortant 
environmental concerns related to 
stratospheric ozone depletion, 
perturbations in the chemical 
composition of the atmosphere, 
and climatic changes from c louds, 
aerosols, and greenhouse gases. 
Numerous state-of-the-art instru- 
ments were fiown successfully, and 
Significant data were collected for 
stratospheric and tropospheric 
research. For stratospheric experi- 
ments, the following are included: 
the Meteorological Measurement 
System (MMS), which measures 
pressure, temperature, and the 
wind vector; and a dual -channel 
tracer instrument for stratospheric 
dynamics studies (Argus). 

These instruments were flown 
on the ER-2 aircraft for the Airborne 
Southern Hemisphere Ozone 
Experiment and for the Stratospheric 
Tracers Atmospheric Transport 
(STRAT) study. For trof)osf)heric 
experiments, instruments were flown 
on several aircraft to measure carbon 
monoxide, nonmethane hydrocar- 
bons, carbonyls, and reactive 
nitrogen species. Specifically, 



145 



fluxes of important biogenic gases 
from terrestrial ecosystems were 
quantified, as were contrails of 
subsonic aircraft and cloud effects 
(SUCCESS) and tropospheric aerosol 
radiative forcing (TARFOX). 

Technology developments were 
realized on two fronts— modeling 
and instruments. Model develop- 
ment efforts or simulations based on 
developed models are presented for 
the following phenomena: the effect 
of gravity waves generated by 
convection or stratospheric circula- 
tion; the effects of climate and 
landcover interactions in the boreal 
biome; soil/atmosphere exchange 
of key trace gas species; biomass 
combustion and pyrogenic trace gas 
emissions; seasonal differentiation 
of microbial methane sources; a 
knowledge-based, interactive, 
graphical tool; use of supercomputer 
technology in ecosystem simulation; 
and paleoenvironment studies using 
pollen data and leaf area develop- 
ment. Newly developed instruments 
include: a miniaturized, lightweight, 
tunal)le diode laser spectrometer; an 
MMS on the NASA DC-8 aircraft to 
provide science-quality state vari- 
ables and wind data; an airborne 
disaster-assessment sensor; a digital 
array scanning interferometer; 
and a fully automated, 14-channel 
sunphotometer to fly on remotely 
piloted aircraft and other platforms. 
Also, a new program was created 
(ERAST) to focus critical technology 
development and flight demonstra- 
tion for remotely piloted vehicles. 



Radiation research focuses on 
phenomena associated with the 
interaction of solar radiation with the 
atmosphere and solar system bodies. 
The high-resolution infrared spec- 
troscopy is devoted to basic experi- 
mental and theoretical research into 
the absorption of radiation by gases. 
The purpose is to determine the 
molecular spectroscopic parameters 
needed for the design and interpreta- 
tion of field measurement programs 
related to the environment of Earth's 
and other planetary and stellar 
atmospheres. This report also 
highlights experiments conducted 
with the Radiation Measurement 
System and with a new within-ieaf 
radiative transfer model. 

Using MTPE data and technol- 
ogy, commercial firms can expand 
their businesses and public-sector 
managers can exercise stewardship 
of the nation's natural resources. 
The enabling, productive use of 
MTPE science and technology is 
exemplified by numerous accom- 
plishments including the effects of 
land-use change on the methods 
and assumptions currently in use to 
estimate the influence of land cover 
in Earth's carbon budgets of the 
northern boreal forest in 
Oregon. Other reported data, 
analyses, and technologies can be 
used by researchers who seek 
answ^ers to key Earth science ques- 
tions and by educators who teach 
Earth sciences. 



146 



M I S S 10 S TO P I A S E T E A R T H E S T E R F R 1 S E 




AIRDAS— Use of Remote 
Sensing for Disaster 
Assessment and 
Management 

James Brass, Vincent Ambrosia, 
Robert Slye 

Each year natural disasters in 
the United States cause an estimated 
20 to 50 billion dollars in damage. 
Fires, earthquakes, floods, hurri- 
canes, and tornadoes destroy 
thousands of structures, devastate 
resources, raise havoc with transpor- 
tation and communication networks, 
and cause loss of life. The manage- 
ment of these events is costly and 
difficult. The consequences of these 
disasters are evident, and they lend 
themselves to monitoring, mapping, 
and assessment by remote sensing. 
The goal of this applied research 
project, funded by NASA's Mission 
to Planet Earth Enterprise and the 
U.S. Forest Service, is to develop, 
demonstrate, and transfer this 
technology to value-added compa- 
nies and the user community. The 
development, based on the Airborne 
Infrared Disaster Assessment System 
(AIRDAS), a compact, multispectral 
scanner designed for quick aircraft 
integration and deployment, takes 
advantage of technology developed 
in the Silicon Valley of California 
and government laboratories, to 
collect, analyze, and communicate 
information to disaster management 
personnel. The research was per- 
formed in collaboration with Edward 
Hildum of Sverdrup Technology, 



Robert Higgins of SIMCO Electron- 
ics, Philip Riggan and Robert 
Lockwood of the U.S. Forest Service, 
and Chien Nguyen of Raytheon. 

Recent activity has involved the 
user community, the U.S. Forest 
Service, the California Department 
of Forest and Fire Protection, the Los 
Angeles County Fire Department, the 
Office of Emergency Services, and 
the Federal Emergency Management 
Agency to develop system require- 
ments, review present capabilities, 
and make recommendations for 
system improvements. Major 
upgrades to the system, developed to 
fill user-driven requirements, include 
additional onboard real-time display 
capabilities, a real-time downlink 
capability, and improved user 
interface software for data exploita- 
tion. The system continues to be 
integrated on users' aircraft for 
disaster monitoring and assessment. 
Currently, the system resides on 
NASA's Lear 24, but it has also 
flown on P-3s, the Lear 35 and 31, 
the KingAir 200, and a Navajo. 

The system was used in Brazil in 
FY96 to detect and map deforesta- 
tion and fire activity, and to fly 
disaster monitoring missions in 
California, New Mexico, and 
Arizona. The system is currently 
being upgraded to provide higher- 
resolution datasets and to increase 
the telemetry capability. 

Point of Contact: J. Brass 
(650) 604-5232 
brass@mail.arc.nasa.gov 



Bay Area Digital 
Georesource 

Edwin Sheffner, Sheri Dister, 
Don Sullivan 

The Bay Area Digital 
Georesource (BADGER) is a coop- 
erative project among NASA, 
Lockheed Martin Palo Alto Research 
Laboratory, and a nonprofit organi- 
zation called the Bay Area Shared 
Information Consortium (BASIC). 
The goal of BADGER, funded 
primarily by NASA's High- 
Performance Computing and 
Communication Program, is to 
provide NASA data and technology 
toward development of a system that 
will make digital, geospatial data of 
the San Francisco Bay Area available 
over the Internet. BADGER technol- 
ogy will serve as the basis for an 
ongoing service operated by BASIC. 
BADGER contains three opera- 
tional functions, all of which are 
designed to be performed on the 
Internet. First, BADGER provides a 
''clearinghouse" for information on 
the nature, location, and cost of 
geospatial data of the Bay Area. 
Geospatial data contain a distribu- 
tional component, that is, data that 
can be represented on a map. 
Geospatial data include, for 
example, images of the surface of 
the Earth from ground-based cam- 
eras to satellite systems, digitized 
maps showing location of roads, 
parcels, transmission lines, and 
water features, and thematic maps 
of natural hazards or population 
distributions. BADGER allows users 
to locate geospatial data quickly. 
Second, BADGER provides a 
mechanism for public and private 
providers of geospatial data to 



K c {) s Y s T [■: M S c; i k n c k and T k c h n o l o c; v 



147 



market data to users. After a user has 
located data of interest, he or she 
will be able to place an order 
through BADGER that will access 
the dataset on the server of the data 
provider, transmit the data to the 
user's machine, notify the user and 
the provider of completion of the 
transaction, and bill the user for the 
data delivered. Lastly, BADGER 
provides online applications to the 
user, that is, procedures by which a 
user can overlay and manipulate 
layers of geospatial data to generate 
an information product — fundamen- 
tal, online, geographic information 
system functions. 

Project personnel contributed to 
BADGER by development and 
implementation of the BADGER/ 
Internet home page interface; 
development of secure, robust 
transaction protocols that allow 
BADGER users to access data on 
public and private servers; develop- 
ment of software for interactive 
location of parcels using street 
addresses or assessors' parcel 
numbers; and development and 
implementation of specific BADGER 
applications. 

Two applications were demon- 
strated in FY96. The "Notice of 
Intent" application, developed for 
the Santa Clara Valley Water 
District, allows the owner/operator 
of a commercial enterprise to 
comply with a section of the clean 
water act by filing information 
electronically about the nature and 
destination of runoff from the site. 
The application completes a form 
and supplies three images showing 
the location as outlined on a digital 
orthophoto quad (aerial photograph) 
and map sheet, and an image 
showing the classification of the 



surface area of the site as pervious or 
impervious. The application also 
calculates the percentage of land in 
each category. The online applica- 
tion, designed by BADGER and 
operated by BASIC, Is expected to 
Increase compliance and lower costs 
for the public and the agency 
administering the law. The second 
application developed by Ames staff 
allows users to locate a parcel in the 
Bay Area and display it in relation to 
categories of known natural hazards, 
for example, flood, ground motion 
during earthquakes, and wildfire. 
BADGER is one way in which 
NASA data and computational 
technology are being used to serve 
commerce and real user needs, and 
to demonstrate the utility and 
practicality of the Internet. 

Point of Contact: E. Sheffner 
(650) 604-0021 
ejs@gaia.arc.nasa.gov 



Brazil/United States 
Environmental Monitoring 
and Global Change 
Program 

James Brass, Vincent Ambrosia 

This program involves a 
multidlsciplinary team of interna- 
tional researchers performing 
cooperative studies to Identify and 
mitigate the extent and environmen- 
tal effect of deforestation and 
burning within the dry and wet 
tropical forests of Brazil. This 
research is performed in collabora- 
tion with Philip Riggan of the U.S. 
Forest Service; Joao Pereira, Hellosa 
Miranda, and Antonio Miranda of 



the University of Brazilia; Theresa 
Campos of the National Center for 
Atmospheric Research; Robert 
Higglns of SIMCO Electronics; and 
Edward Hlldum of Sverdrup Tech- 
nology. The research is supported by 
NASA, the U.S. Forest Service, the 
U.S. Aid for International Develop- 
ment, the Instituto Brasiliero de Meio 
Ambient e dos Recusos Naturals 
Renovaveis, the Brazilian National 
Space Institute, the Universities of 
Brazilia and Sao Paulo, and the 
National Center for Atmospheric 
Research. 

Since 1992, aircraft campaigns 
have focused on the savannas of the 
central portion of Brazil. Collecting 
remote sensing and trace gas 
measurements, the campaigns have 
documented greenhouse gas genera- 
tion, nutrient movement, and cloud 
condensation nuclei development in 
fires throughout the central Brazilian 
grasslands. The fire data continue to 
document the variability in flaming 
and smoldering conditions within 
the cerrados, with rates of spread 
between 0.1 and 3.0 meters per 
second and smoldering conditions 
lasting from 1 5 minutes to over 
3 hours. Consistency in emission 
factors for nitrogen and carbon 
species continues, depending on the 
fuel type and condition. 

The FY96 campaign focused on 
deforestation for the first time. Data 
from flights of the Airborne Infrared 
Disaster Assessment System over the 
Amazon basin and dry tropical forest 
documented large areas of cutting, 
similar to that of the Pacific North- 
west of the United States. The sensor 
system was able to discriminate 
between the two major types of 
cutting (clearcuts and selective 



148 



E C S Y S T H M S C 1 K N C K A N D T \i C W N L (i V 



M I S S I OS TO P LAS ET E ARTH E S TE R P R ! S E 



harvesting), where individual high- 
value species are removed from the 
site. 

Although the deforestation Is 
important from many standpoints, 
the missions during August and 
September of 1 996 concentrated on 
the impact of harvesting on fire 
occurrence and behavior. Two types 
of burning occur after harvesting of 
timber, if the area is clearcut, the 
resulting fire is intense and leaves a 
visible fire scar that can be observed 
using remote sensing. If the area is 
selectively harvested, fires occurring 
after the cutting are less intense and 
leave no telltale sign of a fire scar 
because the overstory left behind 
obscures the scar evidence. 

The FY96 campaign provided 
the data necessary to determine the 
relationship between harvesting, fire 
activity, and greenhouse gas genera- 
tion. These remotely sensed data are 
being analyzed and will help 
quantify the amount of harvesting 
that is occurring in each ecosystem 
and the resultant fire emissions. This 
information will be used to support 
the ongoing effort in Brazil to 
inventory the amount of greenhouse 
gases being produced by blomass 
combustion. 

Point of Contact: J. Brass 
(650) 604-5232 
jbrass@mail,arc.nasa.gov 



Carnegie/Ames/Stanford 
Approach Model 

Christopher Potter 

The Carnegie/Ames/Stanford 
Approach (CASA) model was 
developed for the study of contem- 



porary patterns and processes of 
ecosystem trace gas exchange 
between the terrestrial biosphere and 
the atmosphere. The CASA model is 
unique in that it now simulates a 
host of trace gas fluxes (carbon 
dioxide (CO2), methane (CH4), 
carbon monoxide (CO), nitrous 
oxide (N2O), and nitric oxide (NO)) 
on a scale that merges remote 
sensing, climate, radiation, vegeta- 
tion, and soils datasets with nutrient 
transformation processes at the level 
of the soil microsite. This research 
was performed in collaboration with 
Pamela Matson of the University of 
California, Berkeley, Peter Vitousek 
of Stanford University, and Eric A. 
Davidson of the Woods Hole 
Research Center, Massachusetts. 
The model has been refined 
over the past year to run on either a 
daily or a monthly time interval to 
simulate seasonal patterns in carbon 
fixation, nutrient allocation, lltterfall, 
soil nitrogen mineralization, CO2, 
CO, N2O, NO emission, and CH4 
and CO uptake in soils. In FY96, a 
new global application of the CASA 
model, the hypothesis that variability 
in the ratio of N20:N0 emissions 
over broad spatial gradients depends 
upon seasonal patterns of soil 
wetting and drying, was examined, it 
appears that gross rates of nitrogen 
mineralization account for major 
differences In gas emissions among 
ecosystem types. This finding has 
advanced the state of nitrogen (N) 
trace gas modeling notably by 
calculating indices of the global 
distribution of N cycling processes 
and estimating trace gas emissions 
based directly on those quantified 
mineralization fluxes. 

The latest version of the CASA 
model predicts a total emission flux 



of about 6 teragramsdO^^ grams) of 
N-N2O per year and 1 teragrams of 
N-NO per year from soils worldwide 
(as illustrated In the figure (see Color 
Plate 21 in the Appendix)), exclusive 
of N fertilizer sources. While tropical 
forest ecosystems contribute over 
50% of global annual N2O emis- 
sions, cultivated areas are potentially 
significant sources on a regional 
basis. The next research goal that 
builds on the CASA foundation is to 
discern the manner by which to 
represent and apply fertilized soWs 
and associated trace gas emissions. 

Point of Contact: C. Potter 

(650)604-6164 

cpotter@mail.arc.nasa.gov 



Digital Array Scanning 
Interferometer 

Steve Dunagan, Philip Hammer 

A scanning Interferometer, the 
digital array scanning interferometer 
(DASI), has been designed and 
fabricated for use in remote sensing 
of the ecosystem. The interferometer 
is based on charged-coupled-device 
sensor technology, providing 
spectral sensitivity over the range 
from 0.4 to 1 .0 micron. Light from 
the ground scene is focused by 
means of an objective lens to an 
entrance slit aperture oriented 
normal to the scanning (flight) 
direction. After passing through the 
slit aperture, the light is collimated 
and divided using polarization optics 
into equal amplitude polarization 
vectors oriented along the ordinary 
and extraordinary axes of a 



K (; s Y s r \i m S c i k n c h 



A N' I) T I-: c 11 N 1. (i V 



149 



Hawaiian Pathfinder DASI Instrument Blocic Diagram 



pressure canister to maintain 1 atm for 
convective coding of electronics 



fast wide 



+5,+12v 




Fig. 1. Hawaiian Pathfinder DASI instrument block diagram. 



Wollaston prism. The two polariza- 
tion vectors are then recombined, 
collimated, and focused to the array 
detector, where an interferogram is 
formed. Ordinary and extraordinary 
rays experience a path length 
variation across the prism aperture. 
At the zero path difference location, 
the rays interfere constructively, 
regardless of wavelength, giving rise 
to a central peak intensity. On either 
side of this center burst wavelength, 
dependent interference gives rise to 



an interferogram identical to the 
output of a scanning Michelson 
interferometer. The interferogram 
dimension (columns) of the array 
may then be Fourier transformed to 
yield the spectra of light originating 
from a ground pixel defined by the 
projection of the slit width and the 
array column width. Rows of the 
array provide the spatial distribution 
of information along the slit. Scan- 
ning the slit in the flight direction 
sweeps out an image hypercube. 



The detector array, consisting of 
240 by 750 pixels, may be config- 
ured along with the optics in several 
different ways to optimize spectral 
bandwidth and emphasize either 
spectral or spatial resolution. The 
array output is digitized at 1 bits. 
The primary design configuration 
provides a spectral range from 0.5 to 
1 .0 micron, with theoretical spatial 
resolution of 4.3 meters in the flight 
direction and 2.5 meters cross track 



150 



E C S Y S T K M S C i E N C K AND T H C II N 1. <i Y 



Mission to P l a s e t Earth E s t e r p r i s s 



(at a 50,000-foot altitude), and 
theoretical spectral resolution of 
13 nanonneters at 0.5 micron and 
50 nanometers at 1 .0 micron. (Note 
that the interferometer samples at 
equal intervals In wavenumber 
space, resulting in a variation of 
wavelength resolution that scales 
approximately with lambda^). 
The instrument has been 
interfaced to a data acquisition 
system and packaged in a pressur- 
ized container for deployment on a 
high-altitude, remotely piloted 
aircraft. The self-contained system 
requires less than 1 50 watts of 
power and weighs less than 10 kilo- 
grams. The data acquisition system 
comprises an industrial board level 
computer, with an Intel Pentium 
100-megahertz processor, protocol 
control information (PCI) bus, and 
256 megabytes of system random- 
access memory for direct memory 
access storage of hypercube data. A 
digital framegrabber and a small 
computer serial interface (SCSI) disk 
controller interface to the main 
computer board through a PCI 
passive backplane. An onboard 
4.3-gigabyte hard disk permits 
storage of up to 1 6 hypercube 
datasets per flight. Data transfer is 
limited by the continuous streaming 
write speed of the hard disk, permit- 
ting the acquisition of 5 to 1 inter- 
ferogram frames per second. A 
schematic diagram of the device is 
shown in the figure. 

Point of Contact: S. Dunagan 
(650) 604-4560 
sdunagan@mail.arc.nasa.gov 



Effect of Landuse on 
Regional Estimates of 
Coniferous Forest Water 
and Carbon Budgets 

Joseph Coughlan, Jennifer Dungan 

Geographic information systems 
(GISs) and ecosystem models are 
common tools that ''scale up" 
measurements from ecological field 
experiments across landscapes, 
continents, and even the entire 
globe. Many predictions about the 
causes and effects of global change 
on the environment employ this 
scaling method in combination with 
remotely sensed data. NASA's field 
experiments in grasslands (the First 
International Satellite Land Surface 
Climatology Project Field Experi- 
ment) and in northern forests (the 
Boreal Ecosystem Atmospheric 
Study) are two examples of the need 
to match the scale of ecological field 
measurements with remote-sensing 
data. 

Remote-sensing data often come 
from coarse resolution (approxi- 
mately 1 kilometer, or 0.63 mile) 
remote-sensing instruments such as 
the National Oceanic and Atmo- 
spheric Administration's (NCAA's) 
advanced very-high-resolution 
radiometer (AVHRR) and, in the 
future, NASA's Earth Observing 
System, The mismatch between the 
size of the plots used in a field 
experiment and the size of the 
landscape units generated with this 
relatively coarse resolution data can 
cause scaling errors and result in an 
overprediction of carbon and water 
cycling rates. Researchers in this 
project are quantifying this unit-size 
error in a set of scaling experiments. 



When moving from the plot scale 
(30 square meters) to climate model 
scale (1 -degree cells), errors of up to 
25 to 30% in regional water use and 
carbon cycling are observed for 
coniferous forests in the western 
United States. 

Additionally, there is usually a 
mismatch between the size and 
shape of the units representing the 
landscape data used to make the 
regional estimates, but many 
researchers are unfamiliar with this 
error source. Project researchers 
have begun to document this CIS 
error and its effect on landscape- 
scale predictions made with ecosys- 
tem models. The two most common 
data layers are remotely sensed 
vegetation represented as a digital 
image of 1 -kilometer-(0. 63-mile-) 
long cells and large polygons 
mapping soil types that the GIS 
translates into a digital format. Unit 
size issues have important implica- 
tions consequences for the use of 
soils data because soil map units 
usually comprise spatial units much 
larger in size than the vegetation 
map units. 

In FY96, a new method was 
developed to control resizing of soils 
and vegetation units used to predict 
plant photosynthesis. This method 
reconstructs the soils data so that 
they are represented with ''support" 
similar to the remotely sensed 
vegetation, the 1 -kilometer 
(0.63-mile) grid cell. This reconstruc- 
tion method enables researchers to 
conduct controlled experiments 
with GIS data at several spatial 
resolutions. 

The method was implemented 
with a federally sponsored soils 
database for the state of Oregon and 



K C f) S Y S T \i M S i: I K N C K A N [) T K C M N I (i V 



151 



a NOAA AVHRR image. The figure 
(see Color Plate 22 in the Appendix) 
shows two maps of soil water 
holding capacity (SWHC) for a 
portion of the state. The top map 
shows the homogeneous polygons 
commonly used to spatially repre- 
sent soil variables. The bottom map 
shows the result of the new method 
for SWHC. Using this new map 
resulted in an underestimation bias 
of 1 1 % In plant photosynthesis 
estimates for Oregon. The fact that 
results can vary significantly using 
the same model and data raises 
serious issues for understanding 
regional and global effects of climate 
change. 

Point of Contact: J. Coughlan 
(650) 604-5689 
josco@gaia.arc.nasa.gov 



community, and to act as a reposi- 
tory for comments from the commu- 
nity on the current and future 
operation of the program. The task Is 
accomplished through the use of a 
Landsat Program home page on the 
internet. The home page was 
initiated in FY95; in FY96, extensive 
revisions and additions were made. 
Updated regularly with news items 
about the program, the page con- 
tains links to similar pages in the 
United States and abroad. Approxi- 
mately half the users who log on to 
the home page are from outside the 
United States. An online question- 
naire was also added in 1996 that 
will allow users to provide specific 
information on the program 
electronically. 

Point of Contact: E. Sheffner 
(650) 604-0021 
ejs@gaia.arc.nasa.gov 



Landsat Program 

Edwin Sheffner 

The Landsat Program is the 
longest running program to collect 
and distribute satellite imagery of 
Earth. The first Landsat satellite was 
launched in 1972, and the next in 
the series, Landsat 7, is scheduled 
for launch in 1 998. In the quarter 
century of operation, the instruments 
on Landsat have transmitted millions 
of Earth Images that have been 
employed by a broad user commu- 
nity for studies of global change, 
resource monitoring and assessment, 
local and regional planning, disease 
control, and education. 

The primary program support 
task at Ames Research Center is to 
get information on Landsat to all 
elements of the Landsat data user 



Leaf Modeling 

Lee F. Johnson, Chris Hiavka, 
Philip D. Hammer, David L. Peterson 

The proper Interpretation of a 
land surface reflectance detected by 
air- or satellite-borne remote-sensing 
instrumentation generally requires 
corrections for distortions caused by 
the intervening atmosphere. If the 
signal originates from a vegetated 
surface, corrections for canopy and 
leaf anatomical structure are also 
needed. Although considerable 
progress has been made in the 
development of models to isolate 
the Influence of atmosphere and 
canopy, less effort has been given to 
distortion due to leaf structure. 



To address this factor, a wlthin- 
leaf radiative transfer model (Leaf 
Experimental Absorptivity Feasibility 
Model, or LEAFMOD) has been 
developed and tested against 
empirical data. The model simulates 
the photon scattering and absorbing 
processes within the leaf, based 
upon solution of the one- 
dimensional radiative transfer 
equation, assuming a slab of leaf 
material with homogeneous optical 
properties. When run in the forward 
mode, LEAFMOD generates the leaf 
bidirectional reflectance and trans- 
mittance profiles given a specified 
leaf thickness, optical characteristics 
(scattering and absorption profiles) of 
leaf constituents, and concentrations 
of leaf constituents. In the inverse 
mode, LEAFMOD computes the total 
within-leaf absorption and scattering 
coefficients from the measured leaf 
thickness, reflectance, and transmlt- 
tance. Inversion experiments with 
real and simulated data demon- 
strated that the model appropriately 
decouples scattering and absorption 
processes within the leaf. 

The model produced absorption 
profiles for fresh leaves, with peaks 
at locations corresponding to the 
major absorption features of water 
and chlorophyll. Using available 
spectral data for grape leaves, the 
magnitude of the absorption coeffi- 
cient profile in the visible region was 
strongly correlated to chlorophyll 
concentration determined from 
conventional laboratory analysis. 
Spectral data from a leaf-stacking 
and drying experiment demonstrated 
that absorption features related to 
various biochemical constituents 
could be identified in the dry-leaf 
absorption profile derived by 
LEAFMOD. 



152 



E C S V S T K M 



S C 1 K N C K A N n T E C H N L V 



M I S S i S TO P L A S E T E A R T H E S T E R F R I S E 



The model can be used to 
evaluate fresh-leaf spectral sensitivity 
to leaf chemistry, and also to 
investigate the influence of both 
leaf chemistry and leaf amount on 
vegetation canopy reflectance. Barry 
Ganapol, University of Arizona, 
collaborated with the Ames investi- 
gators on this project. 

Point of Contact: L. Johnson 
(650) 604-3331 
ljohnson@maiKarc.nasa.gov 



Mapping Northern 
Ecosystems; Applications 
for Circumpolar Methane 
Exchange 

Vern Vanderbilt^ Guillaume Perry, 
Joel Stearn 

The goals of this project are to 
improve estimates of the areal extent 
of northern, high-latitude wetlands 
and to gain insight into the role of 
northern ecosystems in the global 
methane (CH4) budget and, in so 
doing, provide improved estimates 
of global CH4 exchange. 

The role of northern v^etlands 
in global atmosphere/biosphere 
interactions remains an unresolved 
but key factor in projecting climatic 
change in response to increasing 
atmospheric concentrations of 
greenhouse gases, such as carbon 
dioxide and methane. Wetlands 
north of 45°N represent globally 
significant methane sources — the 
organic carbon stored in the peat 
and soil beneath these ecosystems 
exceeds that of any other biome on 
Earth. Since climatic warming is 
predicted to be most dramatic in 



northern high latitudes, microbial 
decomposition processes acting on 
these carbon stores will almost 
certainly lead to increased atmo- 
spheric loading of carbon dioxide 
(CO2) and CH4, thus potentially 
enhancing the greenhouse phenom- 
enon even further. 

The areal representation of 
boreal wetlands, the source areas for 
CH4, is poorly known on regional to 
global scales, particularly the areal 
extent of bogs, fens, and open-water 
lakes and ponds. Various estimates 
of the areal extent of northern 
wetlands differ nearly sevenfold. 
These differences in global invento- 
ries lead to large discrepancies in the 
estimation of overall exchange of 
CH4 from northern sources and sinks 
because emission rates from the 
various individual land covers (fens, 
bogs, open water, wetlands, and 
nonwetlands) can vary by a factor of 
more than 100:1 . 

The analysis of results for the 
Boreal Ecosystem-Atmosphere Study 
(BOREAS) shows that glitter, the 
sometimes visually blinding sparkle 
of sunlight from water surfaces, 
provides a powerful signal uniquely 
identifying inundated areas. These 
BOREAS results demonstrate that 
inundated areas may easily be 
separated from noninundated areas. 
It is also possible to discriminate and 
accurately classify fens, open water 
areas, and forested areas because the 
visually blinding glitter of sunlight 
off ruffled water surfaces provides a 
strongly angular signature reflection 
characteristic of wetlands. Conse- 
quently, these classifications and 
estimates of the areal extent of fens 
and open water areas are signifi- 
cantly more accurate than similar 
estimates based upon other sources 



of remotely sensed data, such as the 
Landsat thematic mapper (TM) 
sensor. These encouraging results 
provide an order of magnitude 
improvement in the accuracy of 
estimates of the areal extent of 
wetlands in the BOREAS study 
region. 

Point of Contact: V. Vanderbilt 
(650) 604-4254 
vvanderbilt@mail.arc.nasa.gov 



Modern Ecosystems 
Research: Effects of 
Increased UV-B Radiation 

Hector L. D'Antoni, J. W. Skiles 

The link among these efforts is 
the process of stratospheric ozone 
depletion and the consequent 
increase of ultraviolet-B radiation 
(UV-B) in the biosphere. Given the 
deleterious effect of excessive UV-B 
in living organisms, it is imperative 
to establish what effects, if any, are 
produced by the current levels of 
UV-B. 

Using alfalfa (Medicago sativa 
L.) as the experimental plant, several 
hypotheses have been tested, 
including: (1 ) production of screen- 
ing pigments is stimulated by UV-B 
radiation; (2) node elongation is 
affected by UV-B radiation; (3) phe- 
nological stages (that is, flowering) 
?(re controlled by UV-B radiation; 
and (4) alfalfa plants grown under 
UV-B exclusion have higher chloro- 
phyll concentration. In contrast with 
most of the work published in this 
area of research, these experiments 
were carried out with sunlight as the 
only source of UV-B radiation. 



E C S V S T !■ M S C I [■ N <; K A N I) T I- C W N L (i V 



153 



Plants that were grown in a growth 
chamber in June were transferred to 
a shed in the field and grown there 
for 6 to 8 weeks. The shed provided 
two chambers with identical envi- 
ronmental conditions except for the 
UV regime. Chamber 1 is fitted with 
a cellulose acetate screen that 
removes 1 5% of the UV band 
irradiance to simulate a preozone- 
depletion environment. Chamber 2 
is fitted with a polyester screen that 
removes the entire UV-B irradiance 
band. The experiments, run the past 
two summers, have validated all four 
hypotheses and led to new questions 
and hypotheses. Data require 
intensive statistical analysis before 
publication. 

A similar effort with the Univer- 
sity of Mar del Plata in Argentina has 
resulted in several outcomes. The 
Antarctic spring-time ''ozone hole" 
often reaches southern South 
America during the austral spring. 
Data show that 1 5 to 1 8 days of the 
three-month season have a low 
stratospheric ozone concentration in 
Ushuaia (southern Argentina), while 
ground-level UV-B records are 
concomitantly high. Negative 
correlation was found between 
stratospheric ozone concentration 
and UV-B irradiance at ground level. 
Negative correlation is higher in 
the wavelength range of 289 to 
307 nanometers, known to damage 
DNA, chlorophyll, indole acetic 
acid, gibberelins, abscisic acid, and 
other important plant growth 
molecules. 

An extended visit to Argentina 
resulted in the reflectance analyses 
and in daily measurements of 
solar irradiance in the 300- to 
750-nanometer range in Mar del 
Plata (38°S) that can be compared to 



those of Moffett Field, at approxi- 
mately the same latitude in the 
Northern Hemisphere. 

This research was performed in 
collaboration with G. R. Daleo, 
S. Burry, M. C. Lombardo, A. C. 
Mayoral, P. Palacio, and M. E. Trivi 
of the University of Mar del Plata, 
Argentina; Roy Armstrong, J. 
Corredor, and J. Morell of the 
University of Puerto Rico 
(Mayaguez); and Jaime Matta of the 
Ponce School of Medicine, Ponce, 
Puerto Rico. 

Point of Contact: H. D'Antoni 

(650)604-5149 

hdantoni@mail.arc.nasa.gov 



Optimizing an Ecosystem 
Model for Use on Parallel/ 
Distributed Processors 

J. W. Skiles, Cathy Schulbach 

The goal of this project is to 
update models/codes based on 
current ecological knowledge with 
state-of-the-art computer simulation 
technology. Such modifications will 
decrease execution time, enabling 
simulation of larger geographical 
areas and/or performance of more 
simulation experiments. Areas of 
emphasis are in ecology and com- 
puter science, with the expectation 
that techniques developed in this 
project will serve as a blueprint for 
the construction of future Earth 
system models that will be run on 
supercomputers showing how 
processes can be coded for distrib- 
uted processing. In addition, this 
project is expected to enhance Ames 
Research Center's expertise in the 



areas of information technology and 
ecosystem science. 

As a demonstration, a grassland 
ecosystem model is being used to 
simulate the Great Plains (GP) of the 
United States, an area of approxi- 
mately 60,000 advanced very-high- 
resolution radiometer pixels. Large 
amounts of data are necessary to 
initialize the model for operation on 
a geographical area of this size. Data 
required include soil types, plant 
community information, and 
weather drivers. A major undertak- 
ing is the establishment of data 
structures whereby the model can 
access correct information for a 
specified location, access stored 
soils data, calculate intermediate 
and state variables, and update 
biomass and soil-water variables on 
a pixel-by-pixel basis across the time 
frame of the simulation. All this is to 
be done while maintaining the 
biological and ecological dynamics 
of the ecosystem(s) being simulated. 

Computer science areas of 
emphasis are: (1) establishing 
protocols for rehosting scalar code 
to vector supercomputers and 
distributed-processing super- 
computers; (2) validating metrics 
for load balancing and processor 
scheduling; (3) using different types 
of ecological model codes (differ- 
ence equation, partial differential 
equations, correlation equation) in 
the distributed environment; and 
(4) using automated tools to aid in 
the rehosting process. 

Although most ecosystem 
models include state-of-the-art 
ecological concepts, they employ 
scalar coding that does not take 
advantage of the power of contem- 
porary computing platforms. 



154 



E C S Y S T K M S C I H N C E AND T K C H N L (i Y 



M I S S } S TO F LA S E T E AR T H E S T E R P R I S E 



Rehosting these codes to super- 
computers entails restructuring the 
code itself and using available 
command-level instructions during 
compilation. 

Because of weather patterns and 
the resultant plant communities on 
the GP, it is possible that some 
locations will have growing plants 
while others will not. This scenario 
sometimes causes a processor on a 
multiprocessor supercomputer to be 
idle while others are processing the 
plant growth part of the code. 
Activity in this project focuses on 
efficiently balancing the load of each 
processor so that none are idle, 
thereby decreasing the execution 
time for the simulation experiment. 

Progress to date includes the 
construction of interpolation algo- 
rithms for missing data, stochastic 
weather generation for 31 locations 
on the GP, and the establishment of 
gaming scenarios for global change 
modeling, including the incorpora- 
tion of increases or decreases in 
temperatures and precipitation and 
the seasonal timing of precipitation 
events. 

Point of Contact: J. Skiles 

(650)604-3614 

jskiles@mail.arc.nasa.gov 



Paleoenvironmental 
Research 

Hector L. D'Antoni 

The goal of this research is to 
calibrate modern pollen data in 
terms of satellite remote-sensing data 
of modern vegetation, produce 
calibrated equations of pollen as a 



function of vegetation, and use these 
equations to ''hindcast" (as opposed 
to forecast) past vegetation from 
fossil pollen sequences obtained 
from stratified sediment deposits. 

Many models used to predict the 
environmental future of Earth are 
driven by remotely sensed data of 
vegetation. Historical remote-sensing 
databases of vegetation reach back 
only a few decades. Such a time 
frame is not sufficient to appreciate 
long-term environmental processes. 
Paleoenvironmental data might help 
solve this problem. However, 
paleoenvironmental data are not 
compatible with remotely sensed 
data and, hence, they are of limited 
use for modeling. 

The efforts in FY96 concentrated 
on the Oregon Pollen Transect, a 
collection of modern pollen samples 
from central Oregon, taken every 
1 miles from Dayville in central 
Oregon to Newport on the Pacific 
coast, and crossing vegetation zones 
of ponderosa pine, western juniper, 
fir and Douglas fir, western hemlock, 
alder, and sitka spruce. A predictive 
equation of remote-sensing vegeta- 
tion indices such as the normalized 
difference vegetation index (NDVI) 
was completed. Vegetation in 
Oregon has changed during the last 
50 years, and in some localities the 
original floristic composition is 
difficult to determine from current 
evidence. To solve this problem, a 
database has been constructed with 
modern pollen data produced 
between 1930 and 1949. These data 
will help to recalibrate predictive 
equations of NDVI and other indices 
and, in turn, these equations will 
assist to hindcast Holocene (that is, 
the last 10,000 years) vegetation 



indices that can be used in predic- 
tive models of environmental 
change. Thus, the data used to 
parameterize these models will span 
over 10,000 years rather than the 
customary few decades and will 
help differentiate long cycle fluctua- 
tions from meaningful trends of 
environmental change. 

Point of Contact: H, D'Antoni 

(650)604-5149 

hdantoni@maiLarc.nasa.gov 



Scientists' Intelligent 
Graphical Modeling 
Assistant 

Jennifer Dungan 

Over the past 30 years, scientific 
software models have played an 
increasingly prominent role in the 
conduct of science. Unfortunately, 
implementation of scientific models 
can be difficult and time-consuming, 
and software engineering support 
available specifically for construct- 
ing scientific models is scant. 
Scientists' Intelligent Graphical 
Modeling Assistant (SIGMA), an 
interactive, knowledge-based, 
graphical software development 
environment, was created at Ames to 
fill this gap. SIGMA is designed to 
help scientists rapidly prototype their 
scientific models by simplifying 
construction, modification, and 
reuse of modeling software, and to 
provide a supportive computational 
environment for exploratory model 
building. 

In FY96, two well-known 
models describing radiative transfer 
in vegetation canopies were added 



E C S Y S T H M S C I h: N C K AND T K C H N I. (i Y 



155 



to the SIGMA knowledge base. 
Sensitivity experiments were con- 
ducted with these models to investi- 
gate the relative importance of 
model variables in a more systematic 
way than is usual with traditional 
programming languages, in addition, 
two graduate students at separate 
universities tested SIGMA over the 
Internet. They were able to navigate 
the data flow graph for the ecosys- 
tem model, query the knowledge 
base for details on variables and 
equations, and build the rudiments 
of a new model. 

Point of Contact: J. Dungan 

(650)604-3618 

jennifer@gata.arc.nasa.gov 



Airborne Natural 
Radionuclide 
Measurements in the 
Development and 
Validation of Global 
Three-Dimensional 
Models 

Mark Kritz, Stefan Rosner, 
Robert Chatfield, Leonard Pfister 

As the sources, lifetimes, and 
atmospheric behavior of natural 
radionuclides (for example, ^Be, ^^P, 
^^^Pb, and ^^^Rn (radon)) are 
generally well understood, compari- 
son of their observed distributions 
w^ith model prediction is finding 
increasing application in the devel- 
opment and validation of the global 
models used to assess global change. 
Such comparisons, the focus of two 



international workshops (December 
1993 and August 1995) sponsored 
by the World Climate Research 
Programme, are a key element in the 
validation of the models to be used 
to assess the possible impact of 
aircraft emissions on the atmosphere 
in NASA's Atmospheric Effects of 
Aviation Program (AEAP). 

Project activity in this area has 
been divided between the acquisi- 
tion and the application of airborne 
radionuclide measurements for this 
purpose. Over the last decade, the 
instrumentation has flown on more 
than 100 flights aboard NASA 
aircraft, including the C-1 30, C-1 41 , 
DC-8, and ER-2, For example, in the 
summer of 1 994, the instrumentation 
was onboard the C-1 41 on 15 flights 
out of Moffett Field, acquiring a 
statistically significant climatology of 
the summertime vertical distribution 
of radon over northern California. 
This dataset found extensive applica- 
tion at the August 1 995 World 
Climate Research Programme 
(WCRP) Workshop, and is currently 
being used in the validation of the 
NASA AEAP core model. 

A more extensive set of similar 
airborne radon measurements was 
obtained over the central and 
western United States in the summer 
of 1 996, flying aboard a NASA Lear 
jet. In addition to radon, measure- 
ments of carbon monoxide (CO), 
carbon dioxide (CO2), and methane 
(CH4) were made in cooperation 
with the National Oceanic and 
Atmospheric Administration (NOAA) 
Climate Monitoring Diagnostics 
Laboratory, Boulder, Colorado, and 
measurements of ozone (O^) were 
made in cooperation with the 
National Center for Atmospheric 
Research Facility, Boulder, Colo- 



rado. The combined dataset will be 
used by NOAA in refining their 
budgets of these carbon species in 
the northern hemisphere, and in the 
development and validation of at 
least two three-dimensional global 
chemical transport models. 

Point of Contact: M. Kritz 
(650) 604-5493 
mkritz@asrc.cestm.albany.edu 



Reactive Nitrogen 
Data from the Upper 
Troposphere and Lower 
Stratosphere 

Hanwant B. Singh, Alakh Thakur, 
Peter Mariani 

NASA has launched a subsonic 
aircraft assessment program (SASS) 
to determine the effect of the current 
and future subsonic fleet on the 
Earth's environment as quantitatively 
as possible, A key SASS objective is 
to assess the effect of nitric oxide 
(NO) emissions from subsonic 
aircraft on atmospheric ozone (O3), 
especially in the region of the upper 
troposphere/lower stratosphere. A 
large number of airborne studies in 
the free troposphere (for example, 
global troposphere experiment (GTE) 
and tropospheric ozone/stratospheric 
ozone (TROPOZ/STRATOZ)) and 
lower stratosphere (for example, the 
Airborne Arctic Stratospheric 
Expedition II) have been performed 
in the last decade, and a sizable 
body of data on reactive nitrogen 
species has been collected from both 
hemispheres. Although these data 
have been, to a degree, analyzed as 
individual datasets, no comprehen- 



156 



C S Y S T H M S C I i' N C !■. A N I) T I- C H N i, (; V 



M I S S I S TO P L A S E T E A X T H E S T £ R P R I S £ 



sive analysis and synthesis has been 
undertaken. Tasks in this study are to 
acquire, analyze, and interpret 
experimental data on reactive 
nitrogen species (along with ozone, 
hydrocarbons, and selected tracers) 
that have been collected over the 
last decade, with focus on the upper 
troposphere and lower stratosphere. 
The aim is to provide an overall 
synthesis of the present understand- 
ing of reactive nitrogen chemistry 
and recommendations for future 
field research, based on a compre- 
hensive analysis of available data. 
Analysis of these data was com- 
pleted in FY96 and the results are 
being finalized. 

Point of Contact: H. Singh 
(650) 604-6769 
hsingh@mail.arc.nasa.gov 



Airborne Autotracking 
Sunphotometry 

Philip B. Russell/ John M. Livingston, 
James Hanratty, Damon Ried, 
Jill Bauman 

Atmospheric aerosols (suspen- 
sions of airborne particles compris- 
ing hazes, smokes, and thin clouds 
In the troposphere and stratosphere) 
play important roles in determining 
regional and global climates, the 
chemical composition of the atmo- 
sphere, and atmospheric transport 
processes. As knowledge has 
advanced in each of these fields, so 
has recognition of the Importance of 
aerosols. As a result, national and 
international bodies have called for 
increased efforts to measure aerosol 
properties and effects, as a means of 



improving predictions of future 
climate, including greenhouse 
warming, ozone depletion, and 
radiation exposure of humans and 
other organisms. 

A fundamental measure of any 
aerosol is the way it attenuates a 
beam of light of various colors (that 
Is, wavelengths). This attenuation Is 
usually described in terms of the 
quantity optical depth. The depen- 
dence of optical depth on light 
wavelength Is the optical depth 
spectrum. 

The Ames airborne 
sunphotometers determine the 
optical depth spectrum of aerosols 
and thin clouds by tracking the sun 
and measuring the (relative) intensity 
of the solar beam in several spectral 
channels. The tracking head of each 
instrument mounts external to the 
aircraft cabin, so as to increase data- 
gathering opportunities relative to 
in-cabin sunphotometers and also to 
avoid data contamination by cabin- 
window effects. Each channel 
consists of a baffled entrance tube, 
an interference filter, a photodlode 
detector, and an integral preampli- 
fier. The fllter/detector/preampllfler 
sets are temperature-controlled to 
avoid thermally induced calibration 
changes. Each instrument Includes 
an entrance-window defogging 
system to prevent condensation (a 
problem otherwise common in 
aircraft descents). Solar tracking is 
achieved by azimuth and elevation 
motors driven by differential sun 
sensors. In general, sun tracking Is 
achieved continuously, independent 
of aircraft pitch, roll, and yaw, 
provided rates do not exceed 
approximately 8 degrees per second 
and the sun Is above the aircraft 



horizon and unblocked by clouds or 
aircraft obstructions (for example, 
tail, antennas). 

Data are digitized and recorded 
by an onboard data acquisition and 
control system. Real-time data 
processing and color display are 
routinely provided. The science 
dataset includes the detector signals, 
derived optical depths, detector 
temperature, sun tracker azimuth 
and elevation angles, tracking errors, 
and time. Each instrument must 
maintain Its radiometric calibration 
(including window and filter trans- 
mlttance, as well as detector 
responslvity and electronic gain) to 
within 1% In each spectral channel 
for periods of several months to a 
year. 

In July 1 996, the first science 
flights of the 14-channel sun- 
photometer (developed under the 
NASA Environmental Research 
Aircraft and Sensor Technology 
Program) were made on the Pelican 
(modified Cessna) aircraft of the 
Center for Interdisciplinary Remotely 
Piloted Aircraft Studies during the 
Tropospheric Aerosol Radiative 
Forcing Observational Experiment. 

Point of Contact: P. Russell 
(650) 604-5404 
prussell@mail.arc.nasa.gov 



Analysis of Stratosphere/ 
Troposphere Exchange 

Leonhard Pfister, Henry Selkirk 

Aircraft atmospheric sampling 
experiments require the support and 
advice of meteorologists to be 
successful. The purpose of this 



A T M S P H K R I C C H K M I S T K V 



157 



project is to analyze meteorological 
fields both during and after field 
experiments so as to maximize the 
scientific return of the sampling 
campaigns. The emphasis in this 
project is to provide meteorological 
satellite information as it pertains 
to stratospheric transport and 
stratosphere/troposphere exchange 
processes. 

During FY96, Ames provided 
meteorological coordination for the 
Stratospheric Tracers of Atmospheric 
Transport (STRAT) ER-2 sampling 
mission and the Tropical Ozone 
Transport Experiment (DC-8 mis- 
sion). Three major STRAT field 
campaigns were undertaken. Back 
trajectory analyses of STRAT data 
yielded evidence that midlatitude 
cirrus shields can determine the 
v^ater vapor content of the lov^/er- 
most stratosphere. Some simple 
cloud height algorithms that can be 
used in real time for flight planning 
for the low-altitude (tropospheric) 
ER-2 flights have been developed. 
These algorithms use a combination 
of meteorological analyses and 
satellite imagery to determine the 
cloud altitude at any point in a 
satellite image. 

Finally, it has been shown that 
the subvisible cirrus observed by the 
DC-8 south of Hawaii are probably 
due to air that originates in 
Micronesia and travels northward to 
the subtropical jet, eastward toward 
Hawaii, and southward — following 
an upper tropospheric/lower strato- 
spheric anticyclone. 

Point of Contact: L. Pfister 

(650)604-3183 

pfister@telsci.arc.nasa.gov 



Use of Argus in 
Atmospheric Studies 

Max Loewenstein 

The Argus instrument, a tunable- 
laser, infrared spectrometer, was 
designed to measure the long-lived 
tracers methane (CH4) and nitrous 
oxide (N2O) in the stratosphere up to 
altitudes of 30 kilometers. The ratio 
of these two tracers as a function of 
altitude provides important informa- 
tion on the dynamics of different 
regions of the stratosphere: the surf 
zone, the tropical pipe region, and 
the polar region distinguished by its 
strong winter vortex isolation. 

In June and September 1996, the 
new Argus lightweight N2O and 
CH4 instrument was completed and 
flown on balloon launches to higher 
than 30 kilometers from Ft. Sumner, 
New Mexico (34 degrees N), the site 
of the NASA National Scientific 
Balloon Facility. In June, the instru- 
ment performed flawlessly in an 
engineering sense, but no reportable 
data resulted because of inadequate 
control of the diode laser mount 
temperatures. This control limitation 
was corrected in time for the Sep- 
tember launch. In September, all 
aspects of the Argus instrument 
performed well, and ascent and 
descent profiles for N2O were 
obtained. The methane channel is 
being brought up to operational 
status. The Argus instrument has been 
prepared and shipped to Juazeiro do 
Norte, Brazil, for use in the Orbiter 
Maneuvering System tropical field 
campaign. 

Point of Contact: M. Loewenstein 
(650) 604-5504 
mloewenstein@mail.arc.nasa.gov 



Airborne Tunable Laser 
Absorption Spectrometer 

Max Loewenstein, James R. Podolske 

The Airborne Tunable Laser 
Absorption Spectrometer (ATLAS) 
instrument was designed and built in 
the mid-1980s to measure strato- 
spheric tracer fields from the ER-2 
high-altitude research aircraft. 
During the decade of its operation, 
ATLAS has been used primarily to 
measure the nitrous oxide (N2O) 
tracer field on several campaigns 
primarily focused on polar and 
midlatitude processes in the strato- 
sphere. N2O is a very long-lived 
tracer of stratospheric dynamics. 
Normally, chaotic-appearing 
observations of molecular constitu- 
ents of the atmosphere such as 
reactive nitrogen (NOy), carbon 
dioxide (CO2), water vapor (H2O), 
chloro-fluoro-carbons (CFG) and 
others are ''smoothed" by construct- 
ing correlation diagrams of these 
observations plotted versus N2O. 
These correlations then provide 
important chemical and dynamical 
diagnostic information on the state 
of the atmosphere. 

The ATLAS instrument took part 
in the continuing Stratospheric 
Tracers of Atmospheric Transport 
(STRAT) campaigns, with field 
deployments carried out in October 
1995 and February, July, and 
September 1996. The instrument 
continues to work flawlessly and 
returns very-high-quality N2O tracer 
data on all flights. ATLAS N2O data 
continue to be the dataset of choice 
for analyses carried out within the 
STRAT community. Data compari- 
sons with the aircraft laser infrared 



158 



A !' M S P H !■: K 1 C C H K M I S T R V 



M I S S I OS TO P LAS E T E ART H E S T E R P Rf S E 



absorption spectrometer (ALIAS) and 
airborne chromatograph atnno- 
spheric trace species (ACATS) N2O 
presented at a July 1 996 STRAT 
workshop show that ATLAS data are 
of the highest quality and are true to 
its stated 3% accuracy. 

During the Airborne Southern 
Hemisphere Ozone Experiment and 
the follow-on Stratospheric Tracers 
of Atmospheric Transport campaign 
in 1995 and 1996, a comprehensive 
new dataset on CO2 and N2O in the 
northern hemisphere extending from 
the equator to 60 degrees N was 
accumulated. Several important 
results have emerged from this new 
and unique dataset: Tropospheric air 
carrying its normal seasonal cycle 
and secular trend of CO2 continu- 
ously enters the lower tropical 
stratosphere and is rapidly (in about 
one month) transported to both 
hemispheres. The mean age of the 
observed midstratospheric air is 
about 6 years. Pollutant exhaust 
gases, which will be deposited by 
proposed new stratospheric aircraft 
into the midlatitude lower strato- 
sphere, will probably exceed the 
best current predictions by 30 to 
100%. These results may have 
important implications for the 
influence of stratospheric aircraft on 
the ozone layer. 

Point of Contact: M. Loewenstein 
(650) 604-5504 
mloewenstein@mail.arc.nasa.gov 



Convectively Generated 
Gravity Waves 

Leonhard Pfister 

This work evaluates the influ- 
ence of gravity waves generated by 
convection on stratospheric circula- 
tion in the tropics. Each convective 
event emits a spectrum of gravity 
waves with a broad variation of 
phase speeds. These waves will 
break at a variety of altitudes 
corresponding to these phase 
speeds; when they do, they will 
exert a force at that altitude that can 
affect the overall circulation of the 
tropical stratosphere. Understanding 
the nature of this force quantitatively 
is essential to obtaining a compre- 
hensive picture of the transport of 
trace constituents such as ozone, as 
well as the constituents that govern 
ozone production and destruction. 
The assessment of the importance of 
these convectively generated gravity 
waves has two steps: evaluation of 
the amplitude and spectrum of 
waves from each convective system 
using measurements, and extrapola- 
tion of these results to global budgets 
using global satellite information. 

This task has progressed in two 
areas. First, half-hourly datasets of 
geostationary operational environ- 
mental satellite (GOES-8) imagery 
for the strongly convective Panama 
region have been prepared for 
spectral analysis. The objective is to 
use the satellite data to establish the 
frequency distribution at shorter 
spatial and time scales than that 
addressed by previous satellite 
spectral studies. Second, two case 
studies of gravity waves generated 
by convection as observed by ER-2 
aircraft measurements have been 



completed. One of these, from the 
Stratospheric Tracers of Atmospheric 
Transport campaign, shows gravity 
waves with horizontal wavelengths 
of about 50 kilometers and vertical 
wavelengths of 5 to 10 kilometers. 
This result suggests that a simple 
''transient mountain at the tropo- 
pause" conceptual model can 
explain only part of the mesoscale 
gravity wave variance. Evidence was 
also found of inertia-gravity waves 
with horizontal wavelengths of 
1000 kilometers, consistent with 
studies that inferred horizontal 
wavelengths using vertical profiles 
and inertia-gravity wave dispersion 
relationships. 

Point of Contact: L. Pfister 

(650)604-3183 

pfister@telsci.arc.nasa.gov 



The ER-2 and DC-8 
Meteorological 
Measurement Systems 

K. Roland Chan, T. Paul Bui, 
Antonio A. Trias, Stuart W. Bowen, 
Jonathan Dean-Day 

The Meteorological Measure- 
ment System (MMS) provides in situ, 
high-resolution meteorological 
parameters (pressure, temperature, 
and the three-dimensional wind 
vector). The MMS consists of three 
major systems: (1 ) an air-motion 
sensing system to measure the 
velocity of the air with respect to 
the aircraft; (2) an aircraft-motion 
sensing system to measure the 
velocity of the aircraft with respect 
to the Earth; and (3) a data acquisi- 
tion system to sample, process, and 



A T M S I' II K R [ C {] \\ [•: M I s r R V 



159 



record the measured quantities. 
Since much of the instrumentation is 
attached to the aircraft at specially 
chosen locations, the MMS is a 
platform-specific instrument that 
cannot be transported from one 
aircraft to another. 

The MMS is uniquely qualified 
to investigate atmospheric mesoscale 
(gravity waves) and microscale 
(turbulence) phenomena. Since the 
MMS provides quality data on 
atmospheric state variables, MMS 
data have been extensively used in 
almost all investigations to interpret 
the in situ experiments aboard the 
same aircraft. 

Over the past decade, the ER-2 
MMS has successfully participated in 
many missions, including the 
Stratospheric Tracers Atmospheric 
Transport missions in FY95 and 
FY96. 

The MMS instrumentation has 
been modified several times. Recent 
significant modifications to improve 
the system reliability include: 
(1) improved navigation data using 
the Litton LN-100G, a lightweight 
ring laser gyro embedded in the 
Global Positioning System inertial 
reference system; (2) improved static 
pressure measurement accuracy and 
precision by using quartz diaphragm 
transducers, which have very low 
thermal expansion coefficients; 
(3) improved pitot pressure measure- 
ment, and hence true-air-speed 
measurement by customizing the 
pressure range calibration to the 
ER-2 flight envelope; and (4) a 
Rosemount open-wired, fast- 
temperature sensor installed forward 
on the fuselage, well outside of 
perturbed boundary layer flow, 



making the temperature measure- 
ment independent of the aircraft 
attitude effects. 

The DC-8 MiMS is a new 
instrument, flight-tested for the first 
time in May 1996. The MMS 
performed very well during the 
subsonic aircraft: contrail and cloud 
effects special study, which also 
began its deployment mission in 
May. A preliminary set of MMS 
products are measurements of static 
pressure (p), temperature (T), the 
east/west component of wind (u, 
with east being positive), the north/ 
south component of wind (v, with 
north being positive), the vertical 
component of v/ind (w, with up 
being positive), latitude, longitude, 
and pressure altitude. Analysis of the 
three sets of temperature data (two 
sets from the T^iow probe and one set 
from the Tfast probe) indicates that 
the self-consistency of the tempera- 
ture measurement is within 
±0.3 degree kelvin of the MMS 
design specifications. 

Point of Contacl: K. Chan 
(650) 604-6263 
rchan@mail.arc.nasa.gov 



Environmental Research 
Aircraft and Sensor 
Technology 

Steve Wegener 

Environmental research aircraft 
and sensor technology (ERAST) 
provides focus for critical technology 
development and flight demonstra- 
tion that reduces the technical and 
economic risk of using remotely 
piloted aircraft (RPA) as a means to 



collect scientific data in a timely and 
cost-effective manner. The ERAST 
sensor element addresses the science 
and payload aspects of RPA mis- 
sions. Early focus centered on 
instruments for high-altitude 
(25 kilometers) atmospheric re- 
search. More recently, efforts have 
supported a wider range of science 
mission development activities, 
embracing a broader range of users. 
The present research is being 
performed in collaboration with the 
National Oceanographic and 
Atmospheric Administration, other 
NASA Centers, the National Center 
for Atmospheric Research, the U.S. 
Department of Energy Laboratories, 
several universities, and the Office of 
Naval Research. New concepts in 
RPA over the horizon (OTH) com- 
munications, key to extended 
missions, are being developed and 
tested. Research in extremely 
miniaturized sensor technologies 
holds promise for profound changes 
in thecapability of RPAs. 

The ERAST sensor element made 
significant contributions to the 
outreach in both the sensor and 
platform communities in FY96. The 
sensor element sponsored three 
retreats for sensor engineers during 
the year, which brought together a 
small group of experienced engi- 
neers from ERAST instrument 
development projects. 

The ERAST sensor element 
demonstrated a major milestone in 
FY96: the ability to monitor and 
command RPA payloads via a 
satellite communications link. The 
airborne data system and an infrared 
camera aboard the Scaled Compos- 
ites D2 RPA were monitored for 
status, the communication package 
was reconfigured in flight, and the 



160 



A T M S P }l H K 1 C C H K M I S T K V 



Mission TO P IAS ET Earth Ester? rise 



camera was commanded to acquire 
imagery and downlink data. This 
accomplishment demonstrates the 
ability to monitor and command 
platforms and payloads OTH, a 
capability critical for operation at 
distances greater than 1 60 kilome- 
ters from homebase. OTH communi- 
cations supporting interactive 
research platforms will enhance 
operational safety and scientific 
productivity. 

The ERAST OTH communica- 
tion development effort provided the 
following OTH hardware capabili- 
ties in FY96: 

• The telemetry interface and data 
system was upgraded to be compat- 
ible with the Tracking and Data 
Relay Satellite System (TDRSS) 
interface electronics. Flight tests 
were conducted in December 1995 
and September 1996. 

• TDRSS Interface electronics. 

• An infrared imaging payload was 
developed to demonstrate the OTH 
capability, build expertise for remote 
sensing science, and foster coopera- 
tion with the Lawrence Livermore 
National Laboratory. 

• System integration and testing 
demonstrated the system capability 
on NASA Lear jet and ERAST D2 
platforms. 

• A systematic approach to 
requesting and obtaining approval 
for use of radio frequency spectra 
for TDRSS demonstration was 
established. 

• A "roadmap" for OTH capability 
for the ACE 2 was established. 
Extremely small sensor systems are 
key to placing meaningful payloads 
at very high altitudes. New 
approaches to sensor systems are 
being explored to reduce payload 



weight, volume, and power 
requirements. 

The ERAST sensor element 
sponsored a review of current 
atmospheric science instruments 
used In the Upper Atmosphere 
Research Program. The intention 
was to identify approaches where 
micromachine concepts might lead 
to significant breakthroughs required 
to shrink, in a revolutionary way, the 
weight, volume, and power con- 
sumption of today's payloads. The 
study (in technical review) con- 
cluded that, although some reduc- 
tion was possible, reductions by a 
factor of ten required new measure- 
ment concepts. New work in surface 
acoustic wave technology suggests 
that sensitivities comparable to 
today's instruments can be achieved 
in shoe-box size packages. In 
collaboration with NASA's Mars 
Pathfinder project, NASA's Sensors 
2000!, Sandia National Laboratories, 
the University of California, Berke- 
ley, and the )et Propulsion Labora- 
tory, the ERAST sensor element is 
funding a laboratory study to 
demonstrate parts-per-blllion 
measurement of carbon monoxide 
(CO), nitrous oxide (N2O), water 
vapor, and ozone with acoustic 
(surface acoustic wave and flexural 
plate) and chemlreslstive technolo- 
gies. Demonstration of these results 
in a modular sensor array instrument 
is slated for FY97. 

Point of Contact: S. Wegener 
(650) 604-6278 
swegener@mail.arc.nasa.gov 



Global Emissions 
Inventories for Radon 
and the Cosmogenic 
Radionuclides 

Mark Kritz 

Natural radionuclide measure- 
ments are finding increasing applica- 
tion In the development and 
validation of global circulation and 
chemical tracer models. In particu- 
lar, comparisons of the atmospheric 
radon distributions predicted by 
global models with radon measure- 
ments made at a surface station and 
aboard NASA aircraft flying in the 
free troposphere have been used In 
two recent World Climate Research 
Programme workshops, and by the 
Global Modeling Initiative compo- 
nent of NASA's Atmospheric Effects 
of Aviation Program, to identify 
shortcomings in the convective and 
long-range horizontal transport 
parameterizations used in state-of- 
the-art global atmospheric models. 
However, as this work has continued 
and progressed and models have 
continued to Improve, there has 
been a growing need to review and 
perhaps reassess current estimates of 
the source functions for these 
radionuclides, including not only 
that of radon (---Rn) but also 
those of ^Be and other natural 
radionuclides. 

To this end, the International 
Global Atmospheric Chemistry 
(IGAC) Program has included a 
natural radionuclide element In its 
Global Emission Inventories Activity 
(GEIA). The natural radionuclide 
panel, functioning under the GEIA/ 
IGAC umbrella, has participants 



A r .M t) s 1' 11 [■: R 1 c [] n |: ,m , s t k v 



161 



from France, Germany, Japan, New 
Zealand, and the United States. 
Projects in FY96, performed in 
collaboration with Stephen Schery 
of the New Mexico Institute of 
Technology and Phil Rasch of the 
National Center for Atmospheric 
Research, included: (1) work on a 
global radon emissions map to 
replace the latom/centimeter^/ 
second global mean value currently 
in widespread use, and (2) initiating 
(as a complement to direct onsite 
local point surface flux measure- 
ments) a subproject to infer the 
global mean radon source functions 
using long-term observed mean 
atmospheric distributions of radon 
and the deposition at the surface of 
its decay product ^^^Pb. 

Point of Contact: M. Kritz 
(650) 604-5493 
mkritz@asrc.cestm.albany.edu 



The Great African Plume: 
Tropical Carbon Monoxide 
and Ozone Simulation 

Robert B. Chatfield 



Maintenance of high concentra- 
tions of pollutants, for example, 
tropospheric carbon monoxide and 
ozone, in seemingly distant regions 
of the tropical troposphere is the 
focus of this research. Each of these 
compounds has a major role in 
controlling the oxidizing capacity of 
the lower atmosphere. It was thought 
that the source of a particularly 
striking accumulation of tropo- 
spheric ozone in the middle of the 
Equatorial Atlantic Ocean was due 
to biomass burning taking place in 



South America or Africa. However, 
until this work, the roles that each 
played and the process for mid- 
oceanic, midtropospheric pollution 
were not clear. The work describes 
the origin and character of the 
Great African Plume from biomass 
burning. 

A three-dimensional tropo- 
spheric chemistry model that 
assimilates the winds and weather 
that prevailed during NASA's 
airborne expedition, the Tropical 
Atmospheric Chemistry Experi- 
pnent— Atlantic, has been created. 
The model also helps to establish 
important quantitative details, such 
as the amount of biomass-burning 
emission of carbon monoxide and 
the ability of clouds to vent material 
to the upper troposphere. The latter 
venting process creates the Upper 
South American Plumes and the 
Upper African Plume, which pollute 
virtually the entire southern 
hemisphere. 

The maintenance of the concen- 
trations of carbon monoxide and 
tropospheric ozone are the results of 
complex interplay of many natural 
and social phenomena that make up 
tropical agriculture, followed by 
interactions of many chemical 
species and meteorological pro- 
cesses. This project has made 
explicit several, while pointing out a 
remaining anomaly in the theory that 
is now widely recognized. The most 
visible concentration of tropospheric 
ozone in the mid-Atlantic appears to 
be caused by the Great African 
Plume, which extends up to about 
6 kilometers. This plume arises out 
of a convergence of polluted mate- 
rial in Equatorial and Southern 
Africa. Pollution is then lofted in the 
interoceanic convergence of the 



region, by processes that involve 
intense boundary layer mixing and 
convective clouds reaching to the 
upper troposphere. The lower 
tropospheric mixing process is the 
main source of slow, westward 
moving pollution episodes that 
produce maxima in tropospheric 
carbon monoxide and ozone just 
south of the Equator, and extend to 
the western tip of South America. 
Thunderstorm mixing of pollutants 
tends to lead to southeastward 
movement of these pollutants, and 
great pinwheeling plumes of carbon 
monoxide and ozone spin out from 
the central convective areas of Africa 
and especially South America, 
particularly above 8 kilometers. The 
model has successfully simulated 
aircraft observations in many details, 
and seems to explain more generally 
the satellite observations of carbon 
monoxide. 

Most ozone in the model 
simulations seems to be made close 
to the source, over the continent, 
only a day or so downwind of the 
agricultural burning that is its source. 
Theory and models suggest that 
nitrogen oxides die out, and ozone 
production must then cease. Obser- 
vation, however, has shown a 
remainder of chemically significant 
concentrations of nitrogen oxides in 
the middle and upper troposphere. It 
is, therefore, suggested that there is a 
''re-NOx-ification'' process that 
reproduces photochemically active 
nitrogen oxides, NO^- 

Point of Contact: R. Chatfield 
(650) 604-5490 
chatfield@clio.arc.nasa.gov 



162 



Atmospheric C h k m i s t r v 



M I S S I OS TO P LAS ET E ARTH E S T E RFR I S E 



Instrument for 
Tropospheric Nitrogen 
Studies 

James R. Podolske 

An open path tunable infrared 
monitor of the atnnosphere 
(OPTIMA) instrument is being 
developed to measure nitric acid 
(HNO3) and nitrogen dioxide (NO2) 
from the NASA DC-8 aircraft. 
Assessing the effect of the current 
fleet of commercial subsonic aircraft 
on the Earth's atmosphere requires 
detailed knowledge of the nitrogen 
chemistry of the upper troposphere 
and lower stratosphere. To date, 
understanding of this problem has 
been hampered by large uncertain- 
ties, both in the abundances of the 
odd-nitrogen reservoir species and in 
the partitioning of reactive nitrogen 
(NOy) between nitric oxide (NO) 
and NO2. Among the nitrogen 
reservoir species, HNO3 is expected 
to be one of the predominant 
compounds. Presently, numerical 
models of the upper trososphere and 
lower-stratosphere region predict 
NOx and NOy partitioning which 
differs greatly from that derived from 
existing measurements. Current 
instrumentation has been shown to 
be inadequate for measuring HNO3 
and NO2 with the speed and accu- 
racy required to advance under- 
standing In this area. 

The OPTIMA instrument uses an 
infrared laser spectrometer coupled 
to an actively aligned multiple pass 
Herriott sampling cavity whose open 
absorption path between the fuse- 
lage and the inner engine pylon 
achieves a free-stream absorption 
path length of 384 meters. To further 
enhance the detection sensitivity of 



this tunable infrared diode laser 
system, high-frequency wavelength 
modulation spectroscopy is 
employed. Detection sensitivity for 
these two gases is expected to be in 
the one-to-ten parts per trillion by 
volume range. In FY96, tests of the 
aircraft wing dynamics required to 
specify the alignment system were 
completed. The preliminary design 
of all the instrument subsystems was 
completed, and the final design 
reached the 50% level. 

Point of Contact: J. Podolske 
(650) 604-4853 
jpodolske@mail.arc.nasa.gov 



Reactive Nitrogen and 
Oxygenated Hydrocarbon 
Measurements during the 
Pacific Exploratory 
Mission 

Hanwant B. Singh, W. Viezee, 
R. Chatfield, Y. Chen, D. Herlth, 
R. Kolye 

The Pacific troposphere is a 
major region of the northern hemi- 
sphere that is relatively free of direct 
influences from man-made pollu- 
tion. Tremendous industrial growth 
in Asia and North America in recent 
decades has the potential to signifi- 
cantly affect this pristine region. 
The Pacific Exploratory Mission-B 
(PEM-B) and -Tropics (PEM-Tropics), 
performed in collaboration with 
L. Salas of the San Jose State Founda- 
tion, are major contributors to the 
NASA Global Troposphere Experi- 
ment, designed to study the state of 
the perturbed and the unperturbed 



atmosphere over the Pacific Ocean. 
These experiments aim to elucidate 
mechanisms by which ozone in the 
troposphere is formed, destroyed, 
and transported. To achieve this aim, 
data from the PEM-B mission were 
extensively analyzed and a new set 
of field measurements involving 
nitrogen, carbon, sulfur, and tracer 
species were performed during a 
five-week PEM-Tropics deployment. 
Major deployment sites for PEM- 
Tropics were California, Fiji, Tahiti, 
Easter Island, and New Zealand. 
Highly sensitive measurements of 
reactive nitrogen species, oxygen- 
ated hydrocarbons, and chemical 
tracers were performed by the Ames 
group by deploying an instrument 
called PANAK (PANs/aldehydes/ 
ketones). 

Point of Contact: H. Singh 
(650) 604-6769 
hsingh@mail.arc.nasa.gov 



Subsonic Aircraft: Contrail 
and Cloud Effects Special 
Study 

Owen B, Toon, Steve Hipskind, 
Duane Allen, Paul Bui, Roland Chan, 
Mike Craig, Guy Ferry, Steve Gaines, 
Warren Gore, Eric Jensen, Joe Jordan, 
Stefan Kinne, Bill McKie, 
Peter Pilewskie, Rudi Pueschel, 
Tony Strawa, Annette Walker 

The subsonic aircraft: contrail 
and cloud effects special study 
(SUCCESS) program was a NASA 
field program that used scientifically 
instrumented aircraft and ground- 
based measurements to investigate 
the effects of subsonic aircraft on 



A T M S P [i K K I C C H H M I S T K V 



163 



contrails, cirrus clouds, and atmo- 
spheric chemistry. SUCCESS had 
several objectives. One was to better 
determine the radiative properties of 
cirrus clouds and contrails so that 
satellite observations can more 
reliably measure their effect on 
Earth's radiation budget. Study 
questions included determining 
how cirrus clouds form, whether 
the exhaust from subsonic aircraft 
presently affects the formation of 
cirrus clouds, and, if the exhaust 
does affect the clouds, whether the 
changes induced are of climatologl- 
cal significance. Several new 
instruments were also developed. 
The study also helped to better 
determine the characteristics of 
gaseous and particulate exhaust 
products from subsonic aircraft and 
their evolution in the region near the 
aircraft. 

To achieve these experimental 
objectives, the NASA DC-8 and T-39 
aircraft were used in situ as sampling 
platforms; the NASA ER-2 aircraft 
was deployed as a remote-sensing 
platform; and the NASA 757 was 
used as a source aircraft for studies 
of contrails and exhaust. 

The experiment was cospon- 
sored by NASA's Subsonic Assess- 
ment Program and the Radiation 
Sciences Program, which are part of 
the overall Aeronautics and Mission 
to Planet Earth programs, respec- 
tively, SUCCESS had well over a 
hundred direct participants from 
several NASA Centers, other govern- 
ment agencies, universities, and 
private research companies. 

The SUCCESS project was 
conducted from the Kansas State 
University airport facilities in Salina, 
Kansas, from April 8, 1996, until 
May 10, 1996, with an extension 



from May 10 until May 15, 1996, at 
NASA's Ames Research Center. 
SUCCESS coordinated with the U.S. 
Department of Energy's (DOE's) 
Atmospheric Radiation Measure- 
ments Program, which operates the 
Cloud and Radiation Testbed (CART) 
site located in Northern Oklahoma. 
In addition to the extensive ground- 
based measurements at the CART 
site, DOE also operated an Egret and 
a Twin Otter aircraft, mostly using 
remote-sensing instruments. 

Point of Contact: S. Hipskind 
(650) 604-5076 
shipskind@maiLarc.nasa.gov 



Stratospheric Tracers of 
Atmospheric Transport 

Stephen Hipskind; Michael Craig 

The primary goal of the Strato- 
spheric Tracers of Atmospheric 
Transport (STRAT) program is the 
measurement of the morphology of 
long-lived tracers and dynamical 
quantities as functions of altitude, 
latitude, and season in order to help 
determine rates for global-scale 
transport and future distributions of 
high-speed civil transport (HSCT) 
exhaust emitted into the lower 
stratosphere. The observations 
obtained will improve understanding 
of broader issues involving transport 
of gases and aerosols in the 
stratosphere. 

A secondary goal of STRAT is 
the further characterization of 
atmospheric photochemistry. As 
shown in earlier airborne cam- 
paigns, measurement of free radicals 
within the context of a sufficiently 



large suite of tracer observations 
provides stringent tests for under- 
standing the processes that control 
ozone photochemistry. The STRAT 
campaign has extended the regions 
and seasons for which such mea- 
surements have been taken. 

The experiment made atmo- 
spheric observations with an array 
of instruments on the NASA ER-2 
aircraft that measured long- and 
short-lived chemical tracers, meteo- 
rological parameters, radical species, 
aerosol particles, and solar radiation. 
The experiment consisted of six field 
campaigns of approximately two to 
three weeks each. The individual 
campaigns consisted of local, 
midlatitude flights out of Ames 
Research Center and tropical flights 
to and from the island of Oahu, 
Hawaii. Included in the February 
1 996 campaign was a coordinated 
flight with the NASA DC-8 aircraft 
during the Tropical Ozone Transport 
ExperimentA^ortex Ozone Transport 
Experiment mission. In the coordi- 
nated flight, the ER-2 made measure- 
ments while flying in the exhaust 
plume of the NASA DC-8 aircraft. 

STRAT was sponsored by 
NASA's Atmospheric Effects of 
Aviation Project, Upper Atmosphere 
Research Program, and Atmospheric 
Chemistry Modeling and Analysis 
Program. 

Approximately 100 participants 
from Ames, other NASA centers, 
other government agencies, universi- 
ties, and private consulting compa- 
nies were involved in this research. 
The ER-2 carried a complement of 
15 scientific instruments, making 
primarily in situ measurements with 
two remote-sensing instruments. 
Overall, 53 flights of the ER-2 used 
more than 260 hours of flight time. 



164 



M S P H V. K ! C 



C 11 t- M I S T K V 



M I S S I S TO P L A \ E T E ART H E S T E R P R I S E 



STRAT has provided a rich 
dataset with seasonal coverage over 
two yearS/ vertical profiles from the 
upper troposphere to the lower 
Stratosphere (33,000 to 70,000 feet), 
and latitudinal coverage from 
approximately 2 degrees South to 
more than 60 degrees North. The 
analysis of these data will increase 
understanding of the transport and 
distribution of emissions into the 
lower stratosphere by future HSCTs. 
The data from the first three deploy- 
ments has been published on 
CD-ROM, 

Point of Contact: S. Hipskind 
(650) 604-5076 
shipskind@mail.arc.nasa.gov 



Tropospheric Aerosol 
Radiative Forcing 
Observational Experiment 

Philip B. Russell, John M. Livingston, 
Wendy Whiting 

The overall goal of the Tropo- 
spheric Aerosol Radiative Forcing 
Observational Experiment (TARFOX) 
is to reduce uncertainties in the 
effects of aerosols on climate by 
determining the direct radiative 
effects, as v^ell as the chemical, 
physical, and optical properties of 
the aerosols carried over the western 
Atlantic Ocean from the United 
States. Other objectives of TARFOX 
are (1 ) to perform a variety of closure 
studies by using overdetermined 
datasets to test the mutual consis- 
tency of measurements and calcula- 
tions of a wide range of aerosol 
properties and effects; and (2) to use 



the results of the closure studies to 
assess and reduce uncertainties in 
estimates of aerosol radiative 
forcing, as well as to guide future 
field programs on this subject. 

An important component of the 
closure studies is tests and improve- 
ments of algorithms that retrieve 
aerosol properties and effects from 
satellite and aircraft radiometers. The 
resulting validated algorithms will 
permit extensions of the TARFOX 
results to other times and locations 
that have aerosol properties similar 
to those of the TARFOX Intensive 
Field Period (IFP). 

NASA Ames contributed to 
TARFOX by supplying the Project 
Scientist, Project Manager, an 
ER-2 aircraft, and two airborne 
sunphotometers. The TARFOX IFP, 
including coordinated measurements 
from four satellites (GOES-8, NOAA- 
14, ERS-2, LANDSAT), four aircraft 
(ER-2, C-1 30, C-1 31 , and a modified 
Cessna), land sites, and ships, was 
conducted July 10-31, 1996. A 
variety of aerosol conditions were 
sampled, ranging from relatively 
clean behind frontal passages to 
moderately polluted, with aerosol 
optical depths exceeding 0.5 at 
midvisible wavelengths. The latter 
conditions included separate 
incidents of enhancements caused 
primarily by anthropogenic sources 
and another incident of enhance- 
ment apparently influenced by 
recent fog processing. Spatial 
gradients of aerosol optical thickness 
were sampled to aid in isolating 
aerosol effects from other radiative 
effects and to more tightly constrain 
closure tests, including those of 
satellite retrievals. 



The combined surface, air, and 
space datasets obtained in TARFOX 
will permit a wide variety of closure 
analyses. ''Internal closure" will be 
assessed by comparing the quantities 
thus derived with those deduced 
from the simultaneous in situ 
measurements of aerosol size 
distribution and chemical composi- 
tion. ''External column closure" will 
be assessed by comparing aerosol 
extinction computed from the in situ 
measurements on the C-1 31 A with 
those derived from the airborne 
sunphotometers, satellite radiom- 
eters, and ER-2 imaging 
spectrometers. 

Such closure analyses will yield 
critically needed assessments and 
should reduce the uncertainties in 
derived values of anthropogenic 
aerosol radiative forcing. The closure 
analyses that use satellite optical 
depth and flux results will provide 
tests and, where necessary, improve- 
ments of satellite retrieval algo- 
rithms. The resulting validated 
algorithms will permit extensions of 
the TARFOX results beyond the 
TARFOX period and to other loca- 
tions dominated by similar aerosols 
(for example, the European Atlantic 
coast). 

Point of Contact: P. Russell 
(650) 604-5404 
prussell@mail.arc.nasa.gov 



A T M s [' H E km; C h !■: M 1 S T H V 



165 



Fine Particle Emissions 
by Aircraft 

Rudolf F. Pueschel, Guy V. Ferry, 
Anthony W. Strawa, Duane Allen 

Collections of aerosols were 
made with the Ames wire impactors 
aboard the NASA DC-8 aircraft 
flying in the exhaust of a Boeing 757 
aircraft during the subsonic aircraft: 
contrail and cloud effects special 
study mission. Morphological 
analysis of scanning electron 
micrograph (SEM) images and 
energy-dispersive analysis of x-rays 
emitted during exposure of the 
particles to electrons in the SEM 
allowed a distinction between 
sulfuric acid (H2SO4) and soot 
aerosols. Size classification and 
integration over size distributions 
of both types of aerosols yielded 
ambient concentrations, particle 
surface, particle volume, and 
particle effective radii. This research 
was performed in collaboration with 
Jindra Goodman of San Jose State 
University, Steve Howard of 
Symtech, Sunita Verma of TMA/ 
Norcal, and Glenn Sachse of the 
NASA Langley Research Center. 
The ambient carbon dioxide 
(CO2) concentrations, determined 
at NASA Langley Research Center, 
were 1 .E-4 to 1 .E-5 times the 
concentrations emitted at the engine 
exhaust plane. Applying these 
dilution factors to the soot and 
H2SO4 aerosols yielded emission 
indices between 0.05 < EIh2S04< 
0.5 grams per kilogram of fuel and 
2.5E-4 < Eisoot < 1 -^E-B grams per 
kilogram of fuel, respectively. 



Relating the H2SO4 emission 
indices to the fuel-S (sulfur) content 
resulted in the sulfur gas-to-H2S04 
particle conversion of between 
9 and 24%. This result is close but 
outside the range of 3 to 1 2% of 
sulfur oxidation that could be 
attributed to hydroxyl radical (OH)- 
limited gas-to-particle conversion. 
ThuS; a mechanism of H2SO4 
generation in aircraft exhaust that 
has yet to be determined appears to 
exist. 

The 1990 fuel use by the 
commercial airline fleet amounted 
tol.3E11 kilograms. If the fuel-S 
content averaged 500 parts per 
million mass, a conversion efficiency 
of 20% would have led to a contri- 
bution by aircraft of 1 .3E7 kilograms 
of H2SO4, or approximately one part 
in 1 .E4 of anthropogenic sulfate ion 
(SO4) from other sources. If an 
average Eisoot= 7.5E-4 grams per 
kilograms of fuel is assumed, the 
annual emission of soot into the 
atmosphere in 1 990 would have 
amounted to 1 .E5 kilograms. Com- 
pared with 24 teragrams of total 
anthropogenic soot emission, it 
follows that aircraft contributions 
would have amounted to one part in 
1 .E5 parts of total soot aerosol. 

Thus, aircraft appear to be only 
a minor source for both H2SO4 and 
soot aerosol. Nevertheless, the 
measurements reported highlight the 
issue of particulate (soot and sul- 
fates) emissions from aircraft. The 
potential for the aircraft emissions of 
water vapor (H2O), sulfur (S), and 
particulates to change the nitrogen 
oxides (NOx) effectiveness to 
atmospheric ozone (O3) changes 



may be sufficiently significant to be 
included in models. 

Point of Contact: R. Pueschel 
(650) 604-5254 
rpueschel@mail.arc.nasa.gov 



FIRE Phase III 

Peter Pilewskie, Warren Gore 

Begun in 1986, the First Interna- 
tional Satellite Cloud Climatology 
Project Regional Experiment (FIRE) 
has been an ongoing multiagency 
program designed to promote the 
development of improved cloud and 
radiation parameterizations for use 
in climate models, and to provide for 
assessment and improvement of 
International Satellite Cloud Clima- 
tology Project products. The strategy 
of FIRE has been to combine model- 
ing activities with satellite, airborne, 
and surface observations. 

The third phase of FIRE, or FIRE 
111, has begun with renewed efforts to 
understand the physical and dynami- 
cal properties of climatically impor- 
tant cloud systems, cirrus, and 
stratiform clouds. Project personnel 
completed the second year's work 
developing new solar spectral 
radiometric instrumentation for 
cloud remote sensing and atmo- 
spheric radiative flux measurements. 
In FY96, progress was made toward 
FIRE III goals by testing a prototype 
solar spectral flux radiometer (SSFR) 
at a ground site during the NASA 
subsonic aircraft: contrail and cloud 
effects special study mission. Aside 
from providing important informa- 
tion on the solar radiative properties 
and effects of cirrus and cirrus 



166 



Atmospheric Physics 



M I S S I OS TO P LA S E T EaRTH E S T E R FR 1 S E 



contrails, the benefit of the surface 
deployment was the expedition of 
the developnnent of the airborne 
SSFR, which will be flown in 1 997. 
The team will have airborne and 
surface capabilities to determine the 
net solar spectral flux at multiple 
levels in the atmosphere, and can 
therefore determine the spectral 
absorption characteristics of the 
clear and cloudy atmosphere. 
Uncertainties in the amount of solar 
radiation absorbed by water vapor 
and clouds continue to be a major 
focus in atmospheric radiative and 
climate studies. 

Point of Contact: P. Pilewskie 
(650) 604-0746 
ppilewskie@mail,arc.nasa.gov 



Laboratory Spectroscopy 
of Carbon Dioxide in 
Support of Planetary 
Atmospheres Research 

Lawrence P. Giver, 
Charles Chackerian, Jr. 

Spectra of planets have absorp- 
tion bands caused by the molecules 
in their atmospheres. Measurements 
of these spectral features can yield 
information about these planetary 
atmospheres, such as the abundance 
of the absorbing molecules and the 
pressure and temperature of the gas. 
Measurement requires knowledge of 
absorption band absolute intensities, 
line positions, pressure broadening 
parameters, and temperature depen- 
dencies. Over the past 30 years, 
measurements of these parameters 
have been made on absorption 



bands that are relevant for planetary 
spectra in the NASA-Ames High 
Resolution Spectroscopy Laboratory. 

In collaboration with D. C. 
Benner of the College of William 
and Mary and NASA Langley 
Research Center, and L. R. Brown 
and }. S. Margolis of the NASA Jet 
Propulsion Laboratory, recent work 
has emphasized intensity measure- 
ments of very weak near-infrared 
carbon dioxide (CO2) bands that are 
important for understanding the 
spectrum of the night side of Venus, 
which has a massive CO2 atmo- 
sphere. The first intensity measure- 
ments of several of these bands were 
made using spectra obtained at 
Ames; an article reporting these 
measurements of bands in the 
7000-per-centimeter region was 
published in the Journal of Molecu- 
lar Spectroscopy. Also, measure- 
ments of the perpendicular band at 
531 5 per centimeter were reported 
at the Seventh Annual Conference 
on Laboratory Research for Planetary 
Atmospheres in Kona, Hawaii, in 
October 1995. Collaboration 
between Ames and the NASA 
Langley spectroscopy group resulted 
in a careful comparison of intensity 
measurements of the band at 
4006 per centimeter; these results 
were presented at the Fifty-first Ohio 
State Symposium on Molecular 
Spectroscopy in June 1996. 

As an extension of prior work 
on the visible spectrum of methane 
(CH4), which is abundant in the 
atmospheres of all the outer planets, 
two spectra obtained at the Kitt Peak 
Observatory have shown that a 
previously unidentified absorption 
line in Saturn's spectrum at 



6584.3 angstroms is a relatively 
isolated methane line. 

Point of Contact: L. Giver 
(650) 604-5231 
giver@hires.arc.nasa.gov 



Near-Infrared Remote 
Sensing of Cloud Liquid 
Water 

Peter Pilewskie, Warren Gore 

The 1995 Arizona Program v^as 
a multiinvestigator field experiment 
aimed at advancing the understand- 
ing of winter storm development, 
morphology, and precipitation in a 
mountainous region of central 
Arizona. From January 1 5 through 
March 1 5, 1 995, a wide range of 
instrumentation was in operation 
in and around the Verde Valley 
southwest of Flagstaff, Arizona, 
including the Ames visible and near- 
infrared spectroradiometers, two 
Doppler radars, an instrumented 
aircraft, a lidar, microwave radiom- 
eters, and other surface-based 
instrumentation focused on the 
analysis of wintertime storms in this 
geographically diverse area. Over 
25 scientists from seven institutions 
took part in the program. 

Of special interest to the Ari- 
zona Program is the interaction of 
topographically induced gravity 
waves with the ambient upslope 
flow. It is hypothesized that these 
waves may augment the upslope- 
forced precipitation that falls 
downwind of this feature onto the 
Mogollon Rim. A major thrust of 
the program is to compare the 



A T M S I' H K K i C 



n V s I c s 



167 



observations of these, and other 
aspects of winter storms, with those 
predicted with a numerical model. 

The primary role of this research 
effort was to provide cloud remote- 
sensing analysis for determining 
cloud water thermodynamic phase, 
cloud thickness, and liquid water 
content (LWC). The unique collec- 
tion of surface and in situ sensors 
provided valuable data toward future 
efforts in determining the relation- 
ships between cloud radiative 
properties and microphysics and 
investigations into the role of clouds 
in climate. Understanding the role of 
clouds in climate is predicated not 
only on determination of the com- 
plex manner in which radiative 
energy is redistributed by clouds but 
also on the hydrological processes 
that regulate the distribution of 
condensed water and water vapor in 
the atmosphere. There are funda- 
mental gaps in the present under- 
standing of the hydrological cycle 
that limit the ability to assess the 
present climate, let alone predict 
future climate, or climate response 
in the presence of increased green- 
house gases. Because the removal 
rate of condensed water depends 
strongly on cloud LWC, determining 
the distribution of LWC and its 
dependence on temperature are 
crucial toward a better understand- 
ing of cloud-climate feedback. 

Analysis of the Arizona Program 
data continued in FY96. A new 
method of analyzing surface-base 
cloud transmission spectra, devel- 
oped during the first year of this 
project, was useful in deriving both 
cloud liquid water as well as water 
vapor. Presently, the retrieval 
method derives the relative propor- 
tions of those quantities. New 



analysis, based on photon path 
length distribution calculations, will 
be necessary to obtain liquid water 
and water vapor in absolute units. 
Nevertheless, the work completed 
during the two years of this project 
went far in developing this new 
remote-sensing technique, and will 
continue in the First International 
Satellite Cloud Climatology Project 
Regional Experiment. Furthermore, 
the instrument developed for this 
study, and the data acquired during 
two major field campaigns, were 
applied to other current problems in 
solar radiative transfer, namely, the 
amount of radiation absorbed in 
clouds, and the amount of radiation 
absorbed in the cloud-free 
atmosphere. 

Point of Contact: P. Pilewskie 
(650) 604-0746 
ppilewskie@mail.arc.nasa.gov 



Quantitative Infrared 
Spectroscopy of Minor 
Constituents of the Earth's 

Atmosphere 

Charles Chackerian, Jr., 
Lawrence P. Giver 

The objectives of this study are 
to obtain quantitative laboratory 
spectroscopic measurements of 
molecular constituents, which are of 
importance in understanding the 
'^health" of the Earth's atmosphere, 
and, in particular, emphasize those 
species that are important for 
understanding stratospheric kinetics 
or are used for long-term monitoring 
of the stratosphere. These measure- 
ments provide: (1 ) line and band 



intensity values needed to establish 
limits of detectability for as-yet- 
unobserved species and quantify the 
abundance of those species that are 
observed; (2) line positions, half 
widths, and pressure-induced shifts, 
which are all needed for remote and 
in situ sensing techniques; and 
(3) data on the basic molecular 
parameters at temperatures and 
pressures appropriate to the real 
atmosphere. 

Previously, oxygen (O2) pressure 
broadened spectra of the nitric oxide 
(NO) vibration-rotation fundamental 
were obtained using a flow system. 
These are the first such data to be 
recorded. These spectra will be 
analyzed for O2 broadening coeffi- 
cients for use in interpreting atmo- 
spheric retrievals of NO abundance. 
High-quality 0.005-per-centimeter 
resolution spectra of nitric acid 
(HNO3) were obtained in the 
1 100- to 2700-per-centimeter 
spectral region. Rovibrational line 
intensities on these spectra will be 
determined to support the Open 
Path Trace Species Measurements in 
the Atmosphere experiment. Analy- 
sis for the electric dipole moment 
function of the ground electronic 
state of the hydroxyl radical (OH) 
molecule was completed using a 
very large set of relative emission 
lines of the Meinel system. A 
definitive and comprehensive 
compilation of rovibrational intensi- 
ties for the carbon monoxide (CO) 
molecule in its various isotopic 
forms was published. 

An experimental feasibility study 
was completed for using magnetic 
rotation spectroscopy (MRS) to 
detect free radical molecular species 
(in situ) in the part-per-tritlion mixing 
ratio range. A quantitative theory 



168 



A T M s p n 1-: h i c Physics 



SUCCESS Irradiance 
Measurements 

Peter Pilewskie, Warren Gore 

A solar spectral flux radiometer 
(SSFR) was deployed at the U.S. 
Department of Energy Southern 
Great Plains Cloud and Radiation 
Testbed during April 1996 as part of 
the NASA subsonic aircraft: contrail 
and cloud effects special study 
(SUCCESS). The SSFR was used to 
measure downwelling solar 
spectral irradiance in the 250- to 
2500-nanometer solar spectral 
region, with 10- to 15-nanometer 
resolution. From April 12 through 
April 29, approximately 
1 8,000 spectra were acquired, 
under a variety of meteorological 
conditions; a significant portion of 
the spectra was favorable to the 
study of cirrus clouds and cirrus 
contrails. Analysis of the solar 
spectral flux dataset will facilitate 
the study of the effects of subsonic 
aircraft on the surface and lower 
atmosphere energy budget. Simulta- 
neous in situ observations by the 
NASA DC-8 will provide the micro- 
physical ground truth for assessment 
of the results. 

To determine radiative proper- 
ties of cirrus and cirrus contrails and 
their net effects on radiative energy 
budgets, a proper understanding of 
the solar radiation spectrum in the 
cloud-free atmosphere is needed. 



A key component of the SUCCESS 
objectives was to obtain a better 
understanding of the spectral 
distribution of solar irradiance 
reaching the surface. First results 
from analysis of SSFR data suggest 
that absorption by water vapor can 
be determined quite accurately, and 
the measured spectra agree with 
model calculations within measure- 
ment uncertainties. These findings 
will help eliminate long-standing 
discrepancies between theory and 
observation and, likewise, are 
crucial to determining the effects of 
contrails on the solar energy budget. 

Point of Contact: P. Pilewskie 
(650) 604-0746 
ppilewskie@mail.arc.nasa.gov 



170 



A T M s p H K R I c Physics 



Mission to P l a s e t Earth E m e r f r i s e 



was developed to analyze the MRS 
results obtained as well as to predict 
instrumental parameters required for 
parts per trillion by volume detection 
of free radical molecular species. A 
multipass reflection cell inside a 
solenoid magnet has been con- 
structed to achieve parts-per-trillion- 
by-volume sensitivity. 

Infrared spectra were obtained 
using an interferometer and a 
25-meter base-path multiple reflec- 
tion cell, in the 1.8-micrometer 
region of the infrared red spectrum 
of nitric oxide for the purpose of 
designing a lidar-based smoke-stack 
detector for this molecule. 

Experiments have been per- 
formed to measure line intensities of 
gaseous water in the 1 -micrometer 
spectral region to help in under- 
standing the apparent anomaly 
associated with absorption of solar 
radiation by clouds in the Earth's 
atmosphere. 

Collaborators in this research 
include Chris Mahon of the Space 
Physics Research Institute, 
P. VaranasI of the State University 
of New York, and D. Cooper of the 
Stanford Research Institute. 

Point of Contact: C. Chackerian 
(650) 604-6300 
chack@hires.arc.nasa.gov 



Stratospheric Transport 

Rudolf F. Pueschel, Guy V. Ferry, 
Anthony W. Strawa, Duane Allen 



The atmospheric impact of 
emissions by aircraft depends on 
how they change the steady-state 
abundance of constituents over 
background atmospheric amounts, 



and on the role they play in atmo- 
spheric photochemical, dynamical, 
and radiative processes. Transport 
plays a role in this scheme, because 
the background concentrations of 
atmospheric constituents vary In 
space. For example, effects on 
column ozone of aircraft-emitted 
nitrogen oxides (NO^) result from 
decreases in the middle and upper 
stratosphere and increases in the 
lower stratosphere and troposphere. 
The balance between these Increases 
and decreases depends on the 
dispersion of the aircraft effluent by 
atmospheric transport processes, as 
well as the transport of long-lived 
species, including ozone, in the 
background atmosphere. Thus, one 
important aspect is that fraction of 
exhaust NO^ that would be trans- 
ported from the heavily traveled 
flight corridors In the lower strato- 
sphere in northern midlatitudes to 
the tropics, where it can be lofted to 
higher altitudes by atmospheric 
circulation. Subsequent transport 
back to midlatitudes would be at 
altitudes where the NO^ from the 
exhaust dominates ozone depletion. 
Of Interest also Is the fraction of 
exhaust that could be transported 
from the flight corridors in the 
northern hemisphere across the 
equator to southern latitudes. 
Transport of exhaust from north to 
south would reduce the concentra- 
tion of this fraction and mitigate the 
effects In the northern hemisphere. 

Soot aerosol measurements were 
taken using the Ames wire impactor 
samples to document barriers to 
Interhemispheric mixing, analogous 
to the stratospheric dispersion of 
radioactive ^"^C tracers after the 
nuclear bomb tests In the 1 950s and 
early 1960s. Because atmospheric 



nuclear bomb testing was terminated 
in the early 1960s, the ^^C data 
permit estimates of interhemispheric 
mixing times that turn out to be 
longer than stratospheric residence 
times by more than one order of 
magnitude. This scenario explains a 
decade-long confinement of aircraft 
soot to the hemisphere Into which it 
has been emitted. 

Although there Is agreement 
between the soot and ^"^C datasets In 
horizontal interhemispheric mixing, 
vertical mixing between the soot 
aerosol emission level (12 kilome- 
ters) and the NASA ER-2 aircraft 
ceiling (20 kilometers) is indicated 
only in the soot measurements. 
Transport from midlatitudes to the 
tropics, with subsequent lofting and 
transport back to midlatitudes at 
higher altitudes, is a possibility for 
vertical mixing. However, since ^^C 
in the northern hemisphere does not 
substantiate the soot results, vertical 
motion due to radiometric forces of 
higher absorbing soot particles that 
could prolong their stratospheric 
residence times Is a more plausible 
explanation. 

This research was performed In 
collaboration with JIndra Goodman 
of San Jose State University, Steve 
Howard of Symtech, and Sunita 
Verma of TMA/Norcal. 

Point of Contact: R. Pueschel 
(650) 604-5254 
rpueschel@mail.arc.nasa.gov 



A i M S [* H K R [ C P H V S I C 



169 



The color plates in this appendix correspond to figure citations that appear in the text. Each caption also provides 
the location of the figure citation. 




Color Plate h Air-tratfic controllers at the Dallas/Fort Worth TRACON using FAST advisories to control arrival 
traffic. (Davis, p. 11) 



111 



A p p h \ I) I \ 




Color Plate 2. F-IS ACTIVE superimposed on aerodynamic flight envelope, (lorgensen, p. 19) 



173 




Color Plate 3. Full Apache AH-64D airframe aerodynamic loading in forward fliglit. (Duque, p. 36) 



174 



A P P E S D I X 



Top 



Top 




Perspective 





Perspective 




Color Plate 4. Vertical velocity contours and flow- 
field streamlines, 0-degree angle of incidence. 
(Stremel p, 36) 



Color Plate 5. Vertical velocity contours and flow- 
field streamlines, 1 5-degree angle of incidence. 
(Stremei p, 36) 



175 




Color Plate 6. Vortex cores (red) extracted from a simulation of a helicopter rotor in forward flifiht. 
(Kenwright, p. 39) 



176 



A P P h A I) I \ 




Color Plate 7. Inter active placement and visualization of streaklines in the Virtual Windtunnel 
permits flow features to he easily found. (Kenwright, p. 42) 



111 



U = 56.6 m/8 





60 m/8 



(a) Average axial 
velocity field, 
U(m/8) 



m/8 



50 



mfe 



o-* 



<U'>AJ = 




1.0 <U'>AJ 



(b) Axial velocity 
Auction field, 
<U>AJ 



0^ 



Color Plate 8, PDV measurements of axial velocity and its root-mean-square 
fluctuation amplitudes in a plane normal to the centedine of a low-speed, turbulent 
jet; the view in each image is upstream along the jet centedine. (McKenzie, p. 44) 



178 



A P P E S D I X 




Color Plate 9. The exVis ImageViewer window (left) shows pressure-sensitive paint data tor the 
OAW~3 oblique all-wing model. The white line segments are slice tools that have been interactively 
placed on the image. The GraphViewer (right) shows x-y plots of the coefficient of pressure versus 
length along the tool for each selection. The highlighted peak (the red marker on the green graph) 
was selected in the GraphViewer, resulting in showing the black spot along the gray tool where this 
value occurs. The third window (top right) is the simple user interface used to access datafiles and 
create ImageViewers and CraphViewers. (Uselton, p. 48) 




Color Plate 10. Noise-level contours measured on 4.7%- 
scale DC-W semi-span model in the Ames 12-Foot 
Pressure Wind Tunnel with 52-element Boeing phased 
microphone array. (Home, p. 56} 



179 




Color Plate 1 L Surface pressure contours for multiple design variable study: (a) original wing; (b) 
modified wing with four design variables; (c) modified wing with five design variables. (Greenman, 
p. 58) 




Color Plate 12. Final mesh and computed pressure contours in the rotor plane. (Biswas, p. 59) 



180 



.-1 F r I \ I) I X 




Color Plate 13. Four different types of grids used in computational fluid dynamics. 
Clockwise from upper left: multiple overlapping zone structured grid for the V-22 tiltrotor 
aircraft; periodic multizone structured grid for turbomachinery; hybrid structured and 
unstructured grid for a helicopter rotor; unstructured tetrahedral grid for the Langley fighter 
(Bryson, p. 61) 



181 




Color Plate 14. Optimized HSR configuration. (Cliff, p. 67) 



182 



A P P E S D I X 




Color Plate 15. Thin cirrus clouds scattered the light of the stars in this rare color photograph of the trail of an 
alpha-Monocerotid passing the star Sirius, captured by the Dutch Meteor Society observer Robert Haas at the 
Alcudia-de~Guadix location in Spain during the meteor outburst of November 22, 1 995. (jenniskens, p. 81) 



183 



9 1st salvo 
H 2r>d salvo 
^ 3rd salvo 

Each salvo 
consists of 
8 stations 




25 
20 

15 I 

10 9 

5 S 

0^ 

-5 

-60 -20 20 

Longitude (degrees) 

Color Plate 16. An illustration of possible network of 24 PASCAL stations on Mars. This network is achievable 
using centrifugal release from a spin-stabilized carrier spacecraft and two propulsive time-of-arrival adjustments. 
Three groups of eight stations each follow three sinusoidal loci and are marked by black circles (1st salvo), red 
squares (2nd salvo), and blue diamonds (3rd salvo). Topography is shown in color contours. (Haberle, p. 88) 



184 



A F P E S D I X 




Color Plate 17. The disk, showing a 
pronounced annular gap at the halo/ 
epicycle frequency resonance. This 
pattern persists for the entire length of 
the experiment, which is on the order 
of a Hubble time. (Smith, p. 91) 



Color Plate 18. Strong, non- 

axisymmetric structure is evident in 
the disk at late times in the 
experiment. This bar is absent when 
the halo remains static. Such offset 
bars are often seen at the centers of 
disk galaxies. (Smith, p. 91) 



Meteoroiogicat Data for the Summer (JJA) Season 




180 160 140 120 100 



80 60 180 160 140 120 100 
West longitude (degrees) 



8 10 12 

Contour values 



14 



16 



18 



20 



Color Plate 1 9. (a) The SAGE II frequency of cloud occurrence from to 100% 
in 5% increments, (h) The HIRS frequency of cloud occurrence from to 30% 
in 1 .5%) increments, (c) The average tropopause height from 27,000 to 67,000 
feet in increments of 2,000 feet, (d) The MLS-derived water vapor overburden 
from 4 to 24 precipitable microns in increments of 1 micron. (Haas, p. 95) 



185 




Color Plate 20. The image of the skull of the female from the Visible Human dataset is visualized on the screen 
of the Immersive WorkBench. Rei Cheng (left) and Muriel Ross are studying the 3-D reconstruction v^ith the aid 
of special glasses. Rei Cheng is demonstrating special gloves that will be used to manipulate instruments and 
pieces of bone or other tissue in the immersive environment. (Ross, p. 119) 



186 



A P P E S D I X 




50 



100 



150 
mg N rn'^yr'^ 



200 



250 



>300 



Color PLite 21. Global patterns of annual (a) N2O and (h) NO emissions from soils, (Potter, p, 149) 



187 




a) 



b) 



Color Plate 22. Top map: Conventional map of soil water holding capacity (SWHC) 
using homogeneous polygons. Bottom map: Map of SWHC developed in this research. 
(Coughlan, p. 152) 



188 



^ U.S GOVERNMENT PRINTING OFFICE: 1997-586-175 



REPORT DOCUMENTATION PAGE 



Form Approved 
OMBNo. 0704-0188 



^b.icreport.ngbu.den tor this co.ec.onoi.n«or.auon.se^ 

gathering^d maintaining the data needed, and compleUng f"^ reviewmg the^w^^^^^^ information Operations and Reports. 1215 JeMerson 

li^rgh°:Ltg;:ri^^^^ — 



1. AGENCY USE ONLY (Leavm blank) 



REPORT DATE 

September 1997 



3. REPORT TYPE AND DATES COVERED 

Technical Memorandum 



4. TITLE AND SUBTITLE 

Research and Technology 1996 



6. AUTHOR(S) 

Ames Investigators 



7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 

Ames Research Center 
Moffett Field, CA 94035-1000 



5. FUNDING NUMBERS 



9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 

National Aeronautics and Space Administration 
Washington, DC 20546-0001 



6. PERFORMING ORGANIZATION 
REPORT NUMBER 



A-976441 



10. SPONSORING/MONITORING 
AGENCY REPORT NUMBER 



NASATM-112195 



(650) 604-5500 or contact person(s) at the end of each article 



12a. DISTRIBUTION/AVAILABILITY STATEMENT 

Unclassified — Unlimited 
Subject Category 99 



12b. DISTRIBUTION CODE 



13. ABSTRACT (Maximum 200 words) 

This report highlights the challenging work accomplished during fiscal year 1996 by Ames research 
scientists, engineers, and technologists. It discusses research and technologies that enable the Information 
Age that expand the frontiers of knowledge for aeronautics and space, and that help to maintain U. S. 
leadership in aeronautics and space research and technology development. The accomplishnients span the 
range of goals of NASAs four Strategic Enterprises: Aeronautics and Space Transportation Technology, 
Space Science, Human Exploration and Development of Space, and Mission to Planet Earth. 

The primary purpose of this report is to communicate knowledge-to inform our stakeholders, custom- 
ers and partners, and the people of the United States about the scope and diversity of Ames' mission, the 
nature of Ames' research and technology activities, and the stimulating challenges ahead. The accomplish- 
ments cited illustrate the contributions that Ames is making to improve the quality of life for our citizens and 
the economic position of the United States in the world marketplace. 



14. SUBJECT TERMS 

Aeronautics, Space transportation, Space sciences. Earth sciences, Life sciences. 
Information technology, Research and technology 



17. SECURITY CLASSIFICATION 
OF REPORT 



Unclassified 



NSN 7540-01-280-5500 



18. SECURITY CLASSIFICATION 
OF THIS PAGE 

Unclassified 



19. SECURITY CLASSIFICAIION 
OF ABSTRACT 



15. NUMBER OF PAGES 

201 



16. PRICE CODE 

A09 



20. LIMITATION OF ABSTRACT 



Standard Form 298 (Rev. 2-89) 

Prescribed by ANSI Std. Z39-18 
298-102